Tutorial 7: Finetuning Lightly Checkpoints

LightlySSL provides pre-trained models on various datasets such as ImageNet1k, ImageNet100, Imagenette, and CIFAR-10. All these models’ weights along with their hyperparameter configurations are available here Benchmarks.

In this tutorial, we will learn how to use these pre-trained model checkpoints to fine-tune an image classification model for the Food-101 dataset using PyTorch Lightning.

Imports

Import the Python frameworks we need for this tutorial. Make sure you have the necessary packages installed.

pip install lightly torchmetrics
import pytorch_lightning as pl
import torch
import torch.nn as nn
from pytorch_lightning.loggers import TensorBoardLogger
from torchvision import transforms

from lightly.transforms.utils import IMAGENET_NORMALIZE

Downloading Model Checkpoint

Let’s use the resnet50 model pre-trained on ImageNet1k using the SimCLR method.

You can browse other model checkpoints at Benchmarks.

checkpoint_url = "https://lightly-ssl-checkpoints.s3.amazonaws.com/imagenet_resnet50_simclr_2023-06-22_09-11-13/pretrain/version_0/checkpoints/epoch%3D99-step%3D500400.ckpt"

Tensorboard Logger

tb_logger = TensorBoardLogger("tb_logs", name="lightly_finetuning")

Configuration

Let’s set the configuration parameters for our experiments.

We use a batch size of 32 and an input size of 128.

We only train for 5 epochs because the focus of this tutorial is on finetuning lightly checkpoints.

learning_rate = 0.001
num_workers = 8
batch_size = 3
input_size = 128

seed = 42
num_train_epochs = 5

# use cuda if possible
device = "cuda" if torch.cuda.is_available() else "cpu"

Setup data augmentations and loaders

For this tutorial, we’ll use the Food-101 dataset of 101 food categories with 101,000 images. For each class, 250 manually reviewed test images are provided as well as 750 training images. On purpose, the training images were not cleaned, and thus still contain some amount of noise. This comes mostly in the form of intense colors and sometimes wrong labels. All images were rescaled to have a maximum side length of 512 pixels.

We will also use some minimal augmentations for the train and test subsets. To learn more about data pipelines in LightlySSL you can refer to Tutorial 1: Structure Your Input and to learn more about the different augmentations and learned invariances please refer to Advanced Concepts in Self-Supervised Learning.

from torch.utils.data import DataLoader
from torchvision.datasets import Food101

num_classes = 101


# Training Transformations
train_transform = transforms.Compose(
    [
        transforms.RandomResizedCrop(input_size),
        transforms.RandomHorizontalFlip(),
        transforms.ToTensor(),
        transforms.Normalize(
            mean=IMAGENET_NORMALIZE["mean"],
            std=IMAGENET_NORMALIZE["std"],
        ),
    ]
)
train_dataset = Food101(
    "datasets/food101", split="train", download=True, transform=train_transform
)

train_dataloader = DataLoader(
    train_dataset,
    batch_size=batch_size,
    shuffle=True,
    drop_last=True,
    num_workers=num_workers,
)

# Validation Transformations
val_transform = transforms.Compose(
    [
        transforms.Resize(input_size),
        transforms.CenterCrop(input_size),
        transforms.ToTensor(),
        transforms.Normalize(
            mean=IMAGENET_NORMALIZE["mean"],
            std=IMAGENET_NORMALIZE["std"],
        ),
    ]
)
val_dataset = Food101(
    "datasets/food101", split="test", download=True, transform=val_transform
)
val_dataloader = DataLoader(
    val_dataset,
    batch_size=batch_size,
    shuffle=True,
    drop_last=True,
    num_workers=num_workers,
)
Downloading https://data.vision.ee.ethz.ch/cvl/food-101.tar.gz to datasets/food101/food-101.tar.gz

  0%|          | 0/4996278331 [00:00<?, ?it/s]
  0%|          | 6127616/4996278331 [00:00<01:21, 61093562.62it/s]
  0%|          | 12255232/4996278331 [00:00<01:27, 57226267.63it/s]
  0%|          | 18022400/4996278331 [00:00<01:40, 49501356.62it/s]
  0%|          | 23101440/4996278331 [00:00<01:49, 45600337.80it/s]
  1%|          | 27754496/4996278331 [00:00<01:51, 44570498.75it/s]
  1%|          | 32899072/4996278331 [00:00<01:46, 46632922.19it/s]
  1%|          | 38731776/4996278331 [00:00<01:38, 50156538.43it/s]
  1%|          | 44367872/4996278331 [00:00<01:35, 51986624.72it/s]
  1%|          | 50167808/4996278331 [00:00<01:31, 53787725.39it/s]
  1%|          | 56000512/4996278331 [00:01<01:29, 55081555.53it/s]
  1%|▏         | 63078400/4996278331 [00:01<01:22, 59771444.21it/s]
  1%|▏         | 69337088/4996278331 [00:01<01:21, 60583194.65it/s]
  2%|▏         | 76546048/4996278331 [00:01<01:16, 63997113.93it/s]
  2%|▏         | 84541440/4996278331 [00:01<01:11, 68756981.71it/s]
  2%|▏         | 91455488/4996278331 [00:01<01:13, 66936088.74it/s]
  2%|▏         | 98172928/4996278331 [00:01<01:19, 61284691.00it/s]
  2%|▏         | 104431616/4996278331 [00:01<01:27, 56092385.48it/s]
  2%|▏         | 110329856/4996278331 [00:01<01:26, 56755152.40it/s]
  2%|▏         | 117145600/4996278331 [00:02<01:21, 59873661.30it/s]
  2%|▏         | 124616704/4996278331 [00:02<01:16, 63952013.07it/s]
  3%|▎         | 131137536/4996278331 [00:02<01:17, 63145155.38it/s]
  3%|▎         | 137527296/4996278331 [00:02<01:18, 61717877.87it/s]
  3%|▎         | 143753216/4996278331 [00:02<01:20, 60288551.17it/s]
  3%|▎         | 150536192/4996278331 [00:02<01:17, 62397753.13it/s]
  3%|▎         | 157122560/4996278331 [00:02<01:16, 63336495.89it/s]
  3%|▎         | 163512320/4996278331 [00:02<01:17, 62018871.21it/s]
  3%|▎         | 170262528/4996278331 [00:02<01:15, 63577804.81it/s]
  4%|▎         | 176652288/4996278331 [00:03<01:17, 62198211.02it/s]
  4%|▎         | 182910976/4996278331 [00:03<01:21, 59092384.00it/s]
  4%|▍         | 188874752/4996278331 [00:03<01:22, 58456493.69it/s]
  4%|▍         | 194772992/4996278331 [00:03<01:23, 57702434.15it/s]
  4%|▍         | 200572928/4996278331 [00:03<01:27, 54881212.21it/s]
  4%|▍         | 206110720/4996278331 [00:03<01:33, 51354161.72it/s]
  4%|▍         | 211320832/4996278331 [00:03<01:34, 50692717.74it/s]
  4%|▍         | 216432640/4996278331 [00:03<01:35, 49959747.48it/s]
  4%|▍         | 221478912/4996278331 [00:03<01:50, 43147912.10it/s]
  5%|▍         | 226983936/4996278331 [00:04<01:43, 46185057.44it/s]
  5%|▍         | 232980480/4996278331 [00:04<01:35, 49852218.05it/s]
  5%|▍         | 238223360/4996278331 [00:04<01:34, 50520530.71it/s]
  5%|▍         | 243728384/4996278331 [00:04<01:31, 51771317.26it/s]
  5%|▍         | 249004032/4996278331 [00:04<01:32, 51493822.69it/s]
  5%|▌         | 254902272/4996278331 [00:04<01:28, 53621054.92it/s]
  5%|▌         | 261783552/4996278331 [00:04<01:21, 57999489.78it/s]
  5%|▌         | 268861440/4996278331 [00:04<01:16, 61757494.64it/s]
  6%|▌         | 275087360/4996278331 [00:04<01:16, 61665944.45it/s]
  6%|▌         | 281313280/4996278331 [00:04<01:21, 57681971.99it/s]
  6%|▌         | 287178752/4996278331 [00:05<01:27, 54090351.38it/s]
  6%|▌         | 292716544/4996278331 [00:05<01:26, 54426807.92it/s]
  6%|▌         | 299401216/4996278331 [00:05<01:21, 57855439.79it/s]
  6%|▌         | 306937856/4996278331 [00:05<01:14, 62800664.12it/s]
  6%|▋         | 313819136/4996278331 [00:05<01:12, 64523588.55it/s]
  6%|▋         | 320339968/4996278331 [00:05<01:15, 62012105.78it/s]
  7%|▋         | 326598656/4996278331 [00:05<01:18, 59699156.77it/s]
  7%|▋         | 332627968/4996278331 [00:05<01:23, 55852522.16it/s]
  7%|▋         | 338821120/4996278331 [00:05<01:21, 57463434.64it/s]
  7%|▋         | 345931776/4996278331 [00:06<01:15, 61227362.59it/s]
  7%|▋         | 353206272/4996278331 [00:06<01:12, 64477784.56it/s]
  7%|▋         | 359727104/4996278331 [00:06<01:11, 64619113.34it/s]
  7%|▋         | 366247936/4996278331 [00:06<01:15, 61536939.31it/s]
  7%|▋         | 372473856/4996278331 [00:06<01:18, 58908374.60it/s]
  8%|▊         | 378437632/4996278331 [00:06<01:29, 51515190.72it/s]
  8%|▊         | 383778816/4996278331 [00:06<01:31, 50338073.06it/s]
  8%|▊         | 389414912/4996278331 [00:06<01:28, 51877883.23it/s]
  8%|▊         | 395608064/4996278331 [00:06<01:24, 54572822.74it/s]
  8%|▊         | 401244160/4996278331 [00:07<01:23, 54997778.22it/s]
  8%|▊         | 406913024/4996278331 [00:07<01:22, 55468239.41it/s]
  8%|▊         | 412942336/4996278331 [00:07<01:20, 56801483.31it/s]
  8%|▊         | 418676736/4996278331 [00:07<01:21, 56358308.57it/s]
  8%|▊         | 424607744/4996278331 [00:07<01:20, 57123101.44it/s]
  9%|▊         | 430342144/4996278331 [00:07<01:24, 53820944.92it/s]
  9%|▊         | 436273152/4996278331 [00:07<01:22, 55343079.28it/s]
  9%|▉         | 443023360/4996278331 [00:07<01:17, 58821047.83it/s]
  9%|▉         | 449216512/4996278331 [00:07<01:16, 59645344.52it/s]
  9%|▉         | 455245824/4996278331 [00:07<01:17, 58530252.58it/s]
  9%|▉         | 461733888/4996278331 [00:08<01:15, 60345351.96it/s]
  9%|▉         | 467795968/4996278331 [00:08<01:20, 56365590.46it/s]
  9%|▉         | 473497600/4996278331 [00:08<01:21, 55362207.29it/s]
 10%|▉         | 479952896/4996278331 [00:08<01:17, 57921316.13it/s]
 10%|▉         | 486801408/4996278331 [00:08<01:14, 60936255.51it/s]
 10%|▉         | 492961792/4996278331 [00:08<01:13, 60926292.34it/s]
 10%|▉         | 499286016/4996278331 [00:08<01:12, 61603191.08it/s]
 10%|█         | 505479168/4996278331 [00:08<01:17, 57769680.29it/s]
 10%|█         | 511344640/4996278331 [00:08<01:21, 55153430.31it/s]
 10%|█         | 517636096/4996278331 [00:09<01:18, 57301890.37it/s]
 10%|█         | 524124160/4996278331 [00:09<01:15, 59439872.08it/s]
 11%|█         | 530153472/4996278331 [00:09<01:16, 58349375.86it/s]
 11%|█         | 536051712/4996278331 [00:09<01:19, 55923143.23it/s]
 11%|█         | 541687808/4996278331 [00:09<01:19, 56017154.07it/s]
 11%|█         | 547323904/4996278331 [00:09<01:20, 55286924.00it/s]
 11%|█         | 552894464/4996278331 [00:09<01:25, 52105920.30it/s]
 11%|█         | 558170112/4996278331 [00:09<01:35, 46325094.18it/s]
 11%|█▏        | 562954240/4996278331 [00:09<01:35, 46630227.63it/s]
 11%|█▏        | 567738368/4996278331 [00:10<01:46, 41573298.87it/s]
 11%|█▏        | 573865984/4996278331 [00:10<01:34, 46568681.25it/s]
 12%|█▏        | 581435392/4996278331 [00:10<01:21, 54296606.78it/s]
 12%|█▏        | 588251136/4996278331 [00:10<01:15, 58065197.01it/s]
 12%|█▏        | 594378752/4996278331 [00:10<01:15, 58281760.82it/s]
 12%|█▏        | 600342528/4996278331 [00:10<01:17, 56664593.18it/s]
 12%|█▏        | 606568448/4996278331 [00:10<01:15, 58239047.91it/s]
 12%|█▏        | 614203392/4996278331 [00:10<01:09, 63423610.33it/s]
 12%|█▏        | 620855296/4996278331 [00:10<01:08, 64319715.76it/s]
 13%|█▎        | 628097024/4996278331 [00:11<01:05, 66668321.15it/s]
 13%|█▎        | 634814464/4996278331 [00:11<01:14, 58719087.08it/s]
 13%|█▎        | 640909312/4996278331 [00:11<01:16, 56762217.54it/s]
 13%|█▎        | 647331840/4996278331 [00:11<01:14, 58704097.06it/s]
 13%|█▎        | 654671872/4996278331 [00:11<01:09, 62724405.36it/s]
 13%|█▎        | 661061632/4996278331 [00:11<01:10, 61181773.69it/s]
 13%|█▎        | 667942912/4996278331 [00:11<01:08, 63325772.90it/s]
 13%|█▎        | 674365440/4996278331 [00:11<01:08, 62895579.57it/s]
 14%|█▎        | 680722432/4996278331 [00:11<01:13, 58918797.22it/s]
 14%|█▎        | 686718976/4996278331 [00:12<01:13, 58564040.83it/s]
 14%|█▍        | 692748288/4996278331 [00:12<01:12, 59026386.84it/s]
 14%|█▍        | 699400192/4996278331 [00:12<01:10, 61126198.37it/s]
 14%|█▍        | 706248704/4996278331 [00:12<01:07, 63196280.84it/s]
 14%|█▍        | 712605696/4996278331 [00:12<01:08, 62595116.77it/s]
 14%|█▍        | 718897152/4996278331 [00:12<01:10, 60747864.74it/s]
 15%|█▍        | 725024768/4996278331 [00:12<01:11, 59566408.07it/s]
 15%|█▍        | 731021312/4996278331 [00:12<01:13, 58327213.68it/s]
 15%|█▍        | 736886784/4996278331 [00:12<01:20, 53041913.51it/s]
 15%|█▍        | 742457344/4996278331 [00:13<01:19, 53744099.68it/s]
 15%|█▍        | 747929600/4996278331 [00:13<01:19, 53357040.87it/s]
 15%|█▌        | 753336320/4996278331 [00:13<01:19, 53427594.78it/s]
 15%|█▌        | 759037952/4996278331 [00:13<01:17, 54398324.53it/s]
 15%|█▌        | 765952000/4996278331 [00:13<01:12, 58669552.70it/s]
 15%|█▌        | 772276224/4996278331 [00:13<01:10, 59966902.32it/s]
 16%|█▌        | 778567680/4996278331 [00:13<01:09, 60770622.38it/s]
 16%|█▌        | 784695296/4996278331 [00:13<01:12, 58345824.96it/s]
 16%|█▌        | 790593536/4996278331 [00:13<01:13, 57236541.31it/s]
 16%|█▌        | 796360704/4996278331 [00:13<01:21, 51619337.93it/s]
 16%|█▌        | 801865728/4996278331 [00:14<01:19, 52512667.80it/s]
 16%|█▌        | 808583168/4996278331 [00:14<01:14, 56578902.94it/s]
 16%|█▋        | 814743552/4996278331 [00:14<01:12, 57979809.07it/s]
 16%|█▋        | 820641792/4996278331 [00:14<01:13, 56616249.30it/s]
 17%|█▋        | 826376192/4996278331 [00:14<01:15, 55019544.58it/s]
 17%|█▋        | 832339968/4996278331 [00:14<01:13, 56309107.97it/s]
 17%|█▋        | 838336512/4996278331 [00:14<01:12, 57346766.67it/s]
 17%|█▋        | 844103680/4996278331 [00:14<01:14, 55726331.34it/s]
 17%|█▋        | 849707008/4996278331 [00:14<01:17, 53637025.92it/s]
 17%|█▋        | 855113728/4996278331 [00:15<01:18, 52951442.09it/s]
 17%|█▋        | 861175808/4996278331 [00:15<01:15, 54502889.10it/s]
 17%|█▋        | 866942976/4996278331 [00:15<01:14, 55370542.01it/s]
 17%|█▋        | 873070592/4996278331 [00:15<01:12, 57025003.29it/s]
 18%|█▊        | 878804992/4996278331 [00:15<01:14, 55092660.68it/s]
 18%|█▊        | 884736000/4996278331 [00:15<01:13, 56286806.79it/s]
 18%|█▊        | 890601472/4996278331 [00:15<01:12, 56743855.48it/s]
 18%|█▊        | 896303104/4996278331 [00:15<01:14, 55308650.49it/s]
 18%|█▊        | 901873664/4996278331 [00:15<01:16, 53846074.36it/s]
 18%|█▊        | 907280384/4996278331 [00:16<01:21, 49903135.57it/s]
 18%|█▊        | 912359424/4996278331 [00:16<01:36, 42359919.11it/s]
 18%|█▊        | 916815872/4996278331 [00:16<01:36, 42457081.37it/s]
 18%|█▊        | 922288128/4996278331 [00:16<01:29, 45634292.64it/s]
 19%|█▊        | 928317440/4996278331 [00:16<01:22, 49591582.71it/s]
 19%|█▊        | 935624704/4996278331 [00:16<01:12, 56134236.29it/s]
 19%|█▉        | 941490176/4996278331 [00:16<01:11, 56756280.67it/s]
 19%|█▉        | 947945472/4996278331 [00:16<01:08, 58966703.79it/s]
 19%|█▉        | 953942016/4996278331 [00:16<01:11, 56333823.18it/s]
 19%|█▉        | 960462848/4996278331 [00:16<01:08, 58826957.70it/s]
 19%|█▉        | 966426624/4996278331 [00:17<01:08, 58976116.34it/s]
 19%|█▉        | 972816384/4996278331 [00:17<01:06, 60358357.05it/s]
 20%|█▉        | 980582400/4996278331 [00:17<01:01, 65426335.32it/s]
 20%|█▉        | 987168768/4996278331 [00:17<01:01, 64883972.16it/s]
 20%|█▉        | 993689600/4996278331 [00:17<01:02, 64041873.12it/s]
 20%|██        | 1000243200/4996278331 [00:17<01:02, 64439886.67it/s]
 20%|██        | 1006731264/4996278331 [00:17<01:03, 62463641.09it/s]
 20%|██        | 1013022720/4996278331 [00:17<01:03, 62573291.98it/s]
 20%|██        | 1019314176/4996278331 [00:17<01:05, 60564254.94it/s]
 21%|██        | 1025441792/4996278331 [00:18<01:05, 60711231.20it/s]
 21%|██        | 1031536640/4996278331 [00:18<01:11, 55097521.60it/s]
 21%|██        | 1037172736/4996278331 [00:18<01:11, 55166314.23it/s]
 21%|██        | 1043234816/4996278331 [00:18<01:09, 56671751.77it/s]
 21%|██        | 1050148864/4996278331 [00:18<01:05, 60211242.51it/s]
 21%|██        | 1057816576/4996278331 [00:18<01:00, 64881077.37it/s]
 21%|██▏       | 1064501248/4996278331 [00:18<01:00, 65384430.20it/s]
 21%|██▏       | 1071087616/4996278331 [00:18<01:01, 63610038.13it/s]
 22%|██▏       | 1077510144/4996278331 [00:18<01:03, 61432988.20it/s]
 22%|██▏       | 1083703296/4996278331 [00:19<01:07, 57988650.40it/s]
 22%|██▏       | 1089568768/4996278331 [00:19<01:10, 55380880.48it/s]
 22%|██▏       | 1096777728/4996278331 [00:19<01:05, 59907324.42it/s]
 22%|██▏       | 1104543744/4996278331 [00:19<01:00, 64847306.87it/s]
 22%|██▏       | 1111130112/4996278331 [00:19<00:59, 65109251.63it/s]
 22%|██▏       | 1118666752/4996278331 [00:19<00:56, 68038190.50it/s]
 23%|██▎       | 1125548032/4996278331 [00:19<00:57, 67099589.79it/s]
 23%|██▎       | 1132527616/4996278331 [00:19<00:56, 67864088.20it/s]
 23%|██▎       | 1139343360/4996278331 [00:19<01:01, 62614404.87it/s]
 23%|██▎       | 1145700352/4996278331 [00:19<01:05, 59159338.77it/s]
 23%|██▎       | 1151729664/4996278331 [00:20<01:05, 59036154.01it/s]
 23%|██▎       | 1158610944/4996278331 [00:20<01:02, 61737574.30it/s]
 23%|██▎       | 1166573568/4996278331 [00:20<00:57, 66788914.80it/s]
 23%|██▎       | 1173323776/4996278331 [00:20<00:59, 64714426.52it/s]
 24%|██▎       | 1179877376/4996278331 [00:20<01:00, 63011968.71it/s]
 24%|██▎       | 1186234368/4996278331 [00:20<01:01, 62067974.12it/s]
 24%|██▍       | 1192984576/4996278331 [00:20<00:59, 63592581.71it/s]
 24%|██▍       | 1199374336/4996278331 [00:20<00:59, 63400923.89it/s]
 24%|██▍       | 1205764096/4996278331 [00:20<00:59, 63227735.61it/s]
 24%|██▍       | 1212121088/4996278331 [00:21<01:01, 61268466.60it/s]
 24%|██▍       | 1219428352/4996278331 [00:21<00:58, 64622988.58it/s]
 25%|██▍       | 1225949184/4996278331 [00:21<01:00, 62059998.66it/s]
 25%|██▍       | 1232207872/4996278331 [00:21<01:01, 60977845.30it/s]
 25%|██▍       | 1238335488/4996278331 [00:21<01:07, 55301068.85it/s]
 25%|██▍       | 1244168192/4996278331 [00:21<01:06, 56068033.37it/s]
 25%|██▌       | 1249902592/4996278331 [00:21<01:06, 56395816.56it/s]
 25%|██▌       | 1255800832/4996278331 [00:21<01:05, 57059521.23it/s]
 25%|██▌       | 1261568000/4996278331 [00:21<01:10, 52941814.38it/s]
 25%|██▌       | 1267302400/4996278331 [00:22<01:08, 54116762.59it/s]
 25%|██▌       | 1272807424/4996278331 [00:22<01:11, 52132426.75it/s]
 26%|██▌       | 1278672896/4996278331 [00:22<01:08, 53926941.45it/s]
 26%|██▌       | 1284734976/4996278331 [00:22<01:06, 55767387.62it/s]
 26%|██▌       | 1290371072/4996278331 [00:22<01:06, 55473855.28it/s]
 26%|██▌       | 1296039936/4996278331 [00:22<01:06, 55779065.67it/s]
 26%|██▌       | 1302265856/4996278331 [00:22<01:04, 57502066.56it/s]
 26%|██▌       | 1308065792/4996278331 [00:22<01:04, 56868285.21it/s]
 26%|██▋       | 1314062336/4996278331 [00:22<01:03, 57562403.81it/s]
 26%|██▋       | 1319829504/4996278331 [00:22<01:07, 54768878.76it/s]
 27%|██▋       | 1325367296/4996278331 [00:23<01:10, 52323686.69it/s]
 27%|██▋       | 1330642944/4996278331 [00:23<01:14, 49376686.68it/s]
 27%|██▋       | 1335951360/4996278331 [00:23<01:12, 50347221.37it/s]
 27%|██▋       | 1342144512/4996278331 [00:23<01:08, 53574452.03it/s]
 27%|██▋       | 1349320704/4996278331 [00:23<01:02, 58763339.80it/s]
 27%|██▋       | 1355874304/4996278331 [00:23<01:00, 60624555.39it/s]
 27%|██▋       | 1362001920/4996278331 [00:23<00:59, 60667453.27it/s]
 27%|██▋       | 1368129536/4996278331 [00:23<00:59, 60483230.90it/s]
 28%|██▊       | 1374224384/4996278331 [00:23<01:01, 58812946.25it/s]
 28%|██▊       | 1381007360/4996278331 [00:24<00:58, 61391280.44it/s]
 28%|██▊       | 1388511232/4996278331 [00:24<00:55, 65349506.03it/s]
 28%|██▊       | 1395228672/4996278331 [00:24<00:54, 65882827.80it/s]
 28%|██▊       | 1401847808/4996278331 [00:24<00:55, 64393763.50it/s]
 28%|██▊       | 1408335872/4996278331 [00:24<00:58, 61025639.31it/s]
 28%|██▊       | 1414496256/4996278331 [00:24<01:00, 58834306.57it/s]
 28%|██▊       | 1420427264/4996278331 [00:24<01:00, 58763505.06it/s]
 29%|██▊       | 1427374080/4996278331 [00:24<00:57, 61779760.98it/s]
 29%|██▊       | 1434976256/4996278331 [00:24<00:54, 65832973.64it/s]
 29%|██▉       | 1441890304/4996278331 [00:24<00:53, 66764164.75it/s]
 29%|██▉       | 1448607744/4996278331 [00:25<00:55, 64437741.27it/s]
 29%|██▉       | 1455095808/4996278331 [00:25<00:58, 60131600.02it/s]
 29%|██▉       | 1461190656/4996278331 [00:25<01:03, 55797646.16it/s]
 29%|██▉       | 1466892288/4996278331 [00:25<01:06, 53148789.01it/s]
 29%|██▉       | 1472593920/4996278331 [00:25<01:05, 54109293.70it/s]
 30%|██▉       | 1479049216/4996278331 [00:25<01:01, 56971987.75it/s]
 30%|██▉       | 1485701120/4996278331 [00:25<00:58, 59643807.84it/s]
 30%|██▉       | 1491730432/4996278331 [00:25<01:02, 56219123.83it/s]
 30%|██▉       | 1497432064/4996278331 [00:25<01:02, 55635209.71it/s]
 30%|███       | 1503461376/4996278331 [00:26<01:01, 56877229.62it/s]
 30%|███       | 1510604800/4996278331 [00:26<00:57, 61004144.81it/s]
 30%|███       | 1517551616/4996278331 [00:26<00:54, 63437293.35it/s]
 31%|███       | 1524039680/4996278331 [00:26<00:54, 63797515.06it/s]
 31%|███       | 1531478016/4996278331 [00:26<00:51, 66851701.23it/s]
 31%|███       | 1538228224/4996278331 [00:26<00:51, 66987600.56it/s]
 31%|███       | 1544945664/4996278331 [00:26<00:53, 64982943.23it/s]
 31%|███       | 1551499264/4996278331 [00:26<00:58, 58989560.78it/s]
 31%|███       | 1557528576/4996278331 [00:26<00:59, 57673115.01it/s]
 31%|███▏      | 1563721728/4996278331 [00:27<00:58, 58823521.00it/s]
 31%|███▏      | 1570832384/4996278331 [00:27<00:55, 62226539.83it/s]
 32%|███▏      | 1578303488/4996278331 [00:27<00:51, 65780951.31it/s]
 32%|███▏      | 1584955392/4996278331 [00:27<00:52, 65005482.29it/s]
 32%|███▏      | 1591508992/4996278331 [00:27<00:59, 57506385.12it/s]
 32%|███▏      | 1597440000/4996278331 [00:27<01:00, 56093507.53it/s]
 32%|███▏      | 1603371008/4996278331 [00:27<00:59, 56915748.28it/s]
 32%|███▏      | 1609564160/4996278331 [00:27<00:58, 58199524.54it/s]
 32%|███▏      | 1615462400/4996278331 [00:27<00:58, 57727884.75it/s]
 32%|███▏      | 1621360640/4996278331 [00:28<00:58, 58025125.16it/s]
 33%|███▎      | 1627226112/4996278331 [00:28<00:57, 58202713.01it/s]
 33%|███▎      | 1633091584/4996278331 [00:28<00:58, 57053234.65it/s]
 33%|███▎      | 1638825984/4996278331 [00:28<00:59, 56189164.44it/s]
 33%|███▎      | 1644953600/4996278331 [00:28<00:58, 57605576.78it/s]
 33%|███▎      | 1651113984/4996278331 [00:28<00:56, 58764662.22it/s]
 33%|███▎      | 1657012224/4996278331 [00:28<00:57, 58275768.67it/s]
 33%|███▎      | 1664221184/4996278331 [00:28<00:53, 62272228.88it/s]
 33%|███▎      | 1671659520/4996278331 [00:28<00:50, 65238369.79it/s]
 34%|███▎      | 1678213120/4996278331 [00:28<00:54, 61386253.40it/s]
 34%|███▎      | 1684406272/4996278331 [00:29<00:56, 58893434.20it/s]
 34%|███▍      | 1691090944/4996278331 [00:29<00:54, 61099294.41it/s]
 34%|███▍      | 1697251328/4996278331 [00:29<00:54, 60631841.37it/s]
 34%|███▍      | 1703378944/4996278331 [00:29<00:54, 60295624.50it/s]
 34%|███▍      | 1709899776/4996278331 [00:29<00:53, 61681174.25it/s]
 34%|███▍      | 1716092928/4996278331 [00:29<00:56, 58376376.41it/s]
 34%|███▍      | 1721991168/4996278331 [00:29<00:57, 56825642.97it/s]
 35%|███▍      | 1728184320/4996278331 [00:29<00:56, 58259987.64it/s]
 35%|███▍      | 1734049792/4996278331 [00:29<00:59, 54966088.10it/s]
 35%|███▍      | 1740177408/4996278331 [00:30<00:57, 56672704.17it/s]
 35%|███▍      | 1746305024/4996278331 [00:30<00:56, 57870257.55it/s]
 35%|███▌      | 1752137728/4996278331 [00:30<01:03, 51075000.85it/s]
 35%|███▌      | 1757413376/4996278331 [00:30<01:04, 50504105.81it/s]
 35%|███▌      | 1763213312/4996278331 [00:30<01:01, 52500651.28it/s]
 35%|███▌      | 1769439232/4996278331 [00:30<00:58, 55097470.95it/s]
 36%|███▌      | 1775042560/4996278331 [00:30<01:02, 51535957.44it/s]
 36%|███▌      | 1780449280/4996278331 [00:30<01:01, 52196466.28it/s]
 36%|███▌      | 1785954304/4996278331 [00:30<01:00, 52952078.33it/s]
 36%|███▌      | 1792278528/4996278331 [00:31<00:57, 55823180.97it/s]
 36%|███▌      | 1798045696/4996278331 [00:31<00:56, 56340265.75it/s]
 36%|███▌      | 1804599296/4996278331 [00:31<00:54, 58985346.59it/s]
 36%|███▌      | 1810530304/4996278331 [00:31<00:59, 53288735.22it/s]
 36%|███▋      | 1816002560/4996278331 [00:31<01:02, 50560950.46it/s]
 36%|███▋      | 1821835264/4996278331 [00:31<01:00, 52654295.73it/s]
 37%|███▋      | 1828945920/4996278331 [00:31<00:54, 57793669.26it/s]
 37%|███▋      | 1836384256/4996278331 [00:31<00:50, 62506378.98it/s]
 37%|███▋      | 1842741248/4996278331 [00:31<00:51, 61001742.33it/s]
 37%|███▋      | 1848934400/4996278331 [00:32<00:57, 54428661.17it/s]
 37%|███▋      | 1854570496/4996278331 [00:32<00:57, 54602991.63it/s]
 37%|███▋      | 1860403200/4996278331 [00:32<00:56, 55569867.57it/s]
 37%|███▋      | 1866072064/4996278331 [00:32<01:09, 45363877.22it/s]
 37%|███▋      | 1871609856/4996278331 [00:32<01:05, 47789314.22it/s]
 38%|███▊      | 1877475328/4996278331 [00:32<01:01, 50580023.42it/s]
 38%|███▊      | 1884225536/4996278331 [00:32<00:56, 55140933.11it/s]
 38%|███▊      | 1889992704/4996278331 [00:32<00:56, 54775237.50it/s]
 38%|███▊      | 1895661568/4996278331 [00:32<00:56, 55256570.58it/s]
 38%|███▊      | 1901559808/4996278331 [00:33<00:55, 56228379.85it/s]
 38%|███▊      | 1908408320/4996278331 [00:33<00:51, 59725098.01it/s]
 38%|███▊      | 1916043264/4996278331 [00:33<00:47, 64559978.21it/s]
 38%|███▊      | 1922859008/4996278331 [00:33<00:46, 65617708.13it/s]
 39%|███▊      | 1929478144/4996278331 [00:33<00:48, 63697052.01it/s]
 39%|███▊      | 1935900672/4996278331 [00:33<00:49, 62041623.33it/s]
 39%|███▉      | 1942159360/4996278331 [00:33<00:50, 59886142.93it/s]
 39%|███▉      | 1948614656/4996278331 [00:33<00:49, 61147997.77it/s]
 39%|███▉      | 1954840576/4996278331 [00:33<00:49, 61342746.91it/s]
 39%|███▉      | 1961000960/4996278331 [00:33<00:50, 59901495.45it/s]
 39%|███▉      | 1967620096/4996278331 [00:34<00:49, 61626494.24it/s]
 40%|███▉      | 1973813248/4996278331 [00:34<00:49, 60839733.41it/s]
 40%|███▉      | 1979940864/4996278331 [00:34<00:51, 58809621.94it/s]
 40%|███▉      | 1985871872/4996278331 [00:34<00:52, 57106781.07it/s]
 40%|███▉      | 1992327168/4996278331 [00:34<00:50, 59188640.49it/s]
 40%|████      | 1998913536/4996278331 [00:34<00:49, 61069102.22it/s]
 40%|████      | 2005073920/4996278331 [00:34<00:51, 58557638.50it/s]
 40%|████      | 2010972160/4996278331 [00:34<00:53, 55689395.30it/s]
 40%|████      | 2016608256/4996278331 [00:34<00:54, 54816320.75it/s]
 40%|████      | 2022342656/4996278331 [00:35<00:53, 55496322.36it/s]
 41%|████      | 2027978752/4996278331 [00:35<00:53, 55653233.01it/s]
 41%|████      | 2033582080/4996278331 [00:35<01:00, 48970079.59it/s]
 41%|████      | 2038628352/4996278331 [00:35<01:01, 48143575.27it/s]
 41%|████      | 2044297216/4996278331 [00:35<00:58, 50446287.54it/s]
 41%|████      | 2050523136/4996278331 [00:35<00:54, 53718691.11it/s]
 41%|████      | 2057043968/4996278331 [00:35<00:51, 56916319.35it/s]
 41%|████▏     | 2062843904/4996278331 [00:35<00:54, 53480049.26it/s]
 41%|████▏     | 2068316160/4996278331 [00:35<00:55, 52785124.74it/s]
 42%|████▏     | 2073985024/4996278331 [00:36<00:54, 53835421.78it/s]
 42%|████▏     | 2079981568/4996278331 [00:36<00:52, 55576076.75it/s]
 42%|████▏     | 2086895616/4996278331 [00:36<00:48, 59441840.44it/s]
 42%|████▏     | 2094661632/4996278331 [00:36<00:44, 64766222.79it/s]
 42%|████▏     | 2101739520/4996278331 [00:36<00:43, 66471311.62it/s]
 42%|████▏     | 2108424192/4996278331 [00:36<00:45, 63976198.26it/s]
 42%|████▏     | 2114879488/4996278331 [00:36<00:47, 60640987.18it/s]
 42%|████▏     | 2121302016/4996278331 [00:36<00:46, 61509531.39it/s]
 43%|████▎     | 2128936960/4996278331 [00:36<00:43, 65737229.74it/s]
 43%|████▎     | 2136670208/4996278331 [00:36<00:41, 69057015.60it/s]
 43%|████▎     | 2143649792/4996278331 [00:37<00:41, 68728323.58it/s]
 43%|████▎     | 2151055360/4996278331 [00:37<00:40, 70271495.01it/s]
 43%|████▎     | 2158133248/4996278331 [00:37<00:44, 63476362.48it/s]
 43%|████▎     | 2164621312/4996278331 [00:37<00:46, 60318956.49it/s]
 43%|████▎     | 2170978304/4996278331 [00:37<00:46, 61179828.28it/s]
 44%|████▎     | 2177400832/4996278331 [00:37<00:45, 61988834.63it/s]
 44%|████▎     | 2183692288/4996278331 [00:37<00:45, 61285684.07it/s]
 44%|████▍     | 2190934016/4996278331 [00:37<00:43, 64458263.34it/s]
 44%|████▍     | 2197454848/4996278331 [00:37<00:44, 62460549.98it/s]
 44%|████▍     | 2203779072/4996278331 [00:38<00:44, 62665091.27it/s]
 44%|████▍     | 2210103296/4996278331 [00:38<00:46, 60018050.86it/s]
 44%|████▍     | 2216525824/4996278331 [00:38<00:45, 61199523.45it/s]
 45%|████▍     | 2223833088/4996278331 [00:38<00:42, 64580369.18it/s]
 45%|████▍     | 2230616064/4996278331 [00:38<00:42, 65505137.21it/s]
 45%|████▍     | 2237202432/4996278331 [00:38<00:42, 65463328.97it/s]
 45%|████▍     | 2243788800/4996278331 [00:38<00:42, 64845410.21it/s]
 45%|████▌     | 2250309632/4996278331 [00:38<00:42, 64165041.79it/s]
 45%|████▌     | 2256764928/4996278331 [00:38<00:48, 56244241.02it/s]
 45%|████▌     | 2262695936/4996278331 [00:39<00:47, 56970928.83it/s]
 45%|████▌     | 2268528640/4996278331 [00:39<00:47, 57198799.53it/s]
 46%|████▌     | 2274689024/4996278331 [00:39<00:46, 58299997.84it/s]
 46%|████▌     | 2280620032/4996278331 [00:39<00:46, 58460017.34it/s]
 46%|████▌     | 2286551040/4996278331 [00:39<00:46, 58593721.97it/s]
 46%|████▌     | 2292449280/4996278331 [00:39<00:46, 57625875.02it/s]
 46%|████▌     | 2298249216/4996278331 [00:39<00:47, 57258912.38it/s]
 46%|████▌     | 2304016384/4996278331 [00:39<00:48, 55183835.31it/s]
 46%|████▌     | 2309586944/4996278331 [00:39<00:50, 53144428.15it/s]
 46%|████▋     | 2314960896/4996278331 [00:39<00:51, 52209454.20it/s]
 46%|████▋     | 2320465920/4996278331 [00:40<00:50, 52983215.45it/s]
 47%|████▋     | 2325807104/4996278331 [00:40<00:57, 46846761.04it/s]
 47%|████▋     | 2330624000/4996278331 [00:40<01:00, 43942523.52it/s]
 47%|████▋     | 2336686080/4996278331 [00:40<00:55, 48278710.64it/s]
 47%|████▋     | 2341961728/4996278331 [00:40<00:53, 49472475.56it/s]
 47%|████▋     | 2347204608/4996278331 [00:40<00:52, 50271892.83it/s]
 47%|████▋     | 2353299456/4996278331 [00:40<00:49, 53207988.41it/s]
 47%|████▋     | 2360344576/4996278331 [00:40<00:45, 58154329.86it/s]
 47%|████▋     | 2366537728/4996278331 [00:40<00:44, 59184064.39it/s]
 47%|████▋     | 2373091328/4996278331 [00:41<00:42, 61026618.82it/s]
 48%|████▊     | 2379251712/4996278331 [00:41<00:43, 60584234.07it/s]
 48%|████▊     | 2385412096/4996278331 [00:41<00:42, 60864423.07it/s]
 48%|████▊     | 2391539712/4996278331 [00:41<00:43, 60390890.60it/s]
 48%|████▊     | 2397601792/4996278331 [00:41<00:43, 60278049.18it/s]
 48%|████▊     | 2404319232/4996278331 [00:41<00:41, 62276443.69it/s]
 48%|████▊     | 2412216320/4996278331 [00:41<00:38, 67134925.36it/s]
 48%|████▊     | 2420539392/4996278331 [00:41<00:35, 71868679.57it/s]
 49%|████▊     | 2427813888/4996278331 [00:41<00:35, 72116219.29it/s]
 49%|████▊     | 2435055616/4996278331 [00:41<00:39, 64830343.23it/s]
 49%|████▉     | 2441707520/4996278331 [00:42<00:39, 64102929.82it/s]
 49%|████▉     | 2448228352/4996278331 [00:42<00:45, 55661449.65it/s]
 49%|████▉     | 2454061056/4996278331 [00:42<00:45, 56149505.53it/s]
 49%|████▉     | 2461532160/4996278331 [00:42<00:41, 61048725.70it/s]
 49%|████▉     | 2467823616/4996278331 [00:42<00:42, 59966794.07it/s]
 50%|████▉     | 2473951232/4996278331 [00:42<00:42, 58791359.85it/s]
 50%|████▉     | 2479947776/4996278331 [00:42<00:42, 58766281.94it/s]
 50%|████▉     | 2486009856/4996278331 [00:42<00:42, 59269658.26it/s]
 50%|████▉     | 2492006400/4996278331 [00:42<00:44, 56521061.78it/s]
 50%|████▉     | 2497740800/4996278331 [00:43<00:44, 55970597.29it/s]
 50%|█████     | 2504523776/4996278331 [00:43<00:42, 59320412.62it/s]
 50%|█████     | 2511536128/4996278331 [00:43<00:39, 62375024.18it/s]
 50%|█████     | 2517827584/4996278331 [00:43<00:40, 61407008.61it/s]
 51%|█████     | 2524020736/4996278331 [00:43<00:40, 61390349.45it/s]
 51%|█████     | 2530181120/4996278331 [00:43<00:41, 59936667.65it/s]
 51%|█████     | 2536210432/4996278331 [00:43<00:42, 58321164.92it/s]
 51%|█████     | 2542075904/4996278331 [00:43<00:42, 57630935.87it/s]
 51%|█████     | 2548465664/4996278331 [00:43<00:41, 59402471.08it/s]
 51%|█████     | 2554429440/4996278331 [00:44<00:41, 58632234.19it/s]
 51%|█████     | 2560589824/4996278331 [00:44<00:40, 59441539.44it/s]
 51%|█████▏    | 2566553600/4996278331 [00:44<00:41, 59028063.04it/s]
 51%|█████▏    | 2572484608/4996278331 [00:44<00:43, 56065860.39it/s]
 52%|█████▏    | 2578153472/4996278331 [00:44<00:43, 55699278.45it/s]
 52%|█████▏    | 2584477696/4996278331 [00:44<00:41, 57825033.09it/s]
 52%|█████▏    | 2590310400/4996278331 [00:44<00:42, 56600928.05it/s]
 52%|█████▏    | 2596110336/4996278331 [00:44<00:42, 56753731.14it/s]
 52%|█████▏    | 2602074112/4996278331 [00:44<00:41, 57537641.77it/s]
 52%|█████▏    | 2607841280/4996278331 [00:44<00:44, 54208110.99it/s]
 52%|█████▏    | 2613346304/4996278331 [00:45<00:43, 54386389.37it/s]
 52%|█████▏    | 2618818560/4996278331 [00:45<00:44, 53121206.37it/s]
 53%|█████▎    | 2624159744/4996278331 [00:45<00:48, 49256132.28it/s]
 53%|█████▎    | 2630549504/4996278331 [00:45<00:44, 53241939.03it/s]
 53%|█████▎    | 2636087296/4996278331 [00:45<00:43, 53787387.70it/s]
 53%|█████▎    | 2642313216/4996278331 [00:45<00:41, 56179975.23it/s]
 53%|█████▎    | 2649817088/4996278331 [00:45<00:38, 61615359.45it/s]
 53%|█████▎    | 2656665600/4996278331 [00:45<00:36, 63589851.06it/s]
 53%|█████▎    | 2663088128/4996278331 [00:45<00:38, 61010704.85it/s]
 53%|█████▎    | 2669248512/4996278331 [00:46<00:39, 58614800.34it/s]
 54%|█████▎    | 2675638272/4996278331 [00:46<00:38, 60104890.42it/s]
 54%|█████▎    | 2681995264/4996278331 [00:46<00:37, 61072050.79it/s]
 54%|█████▍    | 2688155648/4996278331 [00:46<00:38, 60678998.84it/s]
 54%|█████▍    | 2695168000/4996278331 [00:46<00:36, 63378990.67it/s]
 54%|█████▍    | 2701557760/4996278331 [00:46<00:37, 61359179.48it/s]
 54%|█████▍    | 2707750912/4996278331 [00:46<00:37, 60524518.46it/s]
 54%|█████▍    | 2713845760/4996278331 [00:46<00:38, 59149568.73it/s]
 54%|█████▍    | 2720268288/4996278331 [00:46<00:37, 60573304.98it/s]
 55%|█████▍    | 2726363136/4996278331 [00:46<00:38, 58568541.42it/s]
 55%|█████▍    | 2732261376/4996278331 [00:47<00:39, 57455104.35it/s]
 55%|█████▍    | 2738290688/4996278331 [00:47<00:38, 58259904.57it/s]
 55%|█████▍    | 2744680448/4996278331 [00:47<00:37, 59767847.85it/s]
 55%|█████▌    | 2750676992/4996278331 [00:47<00:38, 57895975.55it/s]
 55%|█████▌    | 2756509696/4996278331 [00:47<00:40, 55770716.73it/s]
 55%|█████▌    | 2762113024/4996278331 [00:47<00:40, 55582472.66it/s]
 55%|█████▌    | 2767716352/4996278331 [00:47<00:41, 53753148.76it/s]
 56%|█████▌    | 2773516288/4996278331 [00:47<00:40, 54835155.12it/s]
 56%|█████▌    | 2779021312/4996278331 [00:47<00:44, 49999918.39it/s]
 56%|█████▌    | 2784133120/4996278331 [00:48<00:44, 49493812.81it/s]
 56%|█████▌    | 2789146624/4996278331 [00:48<00:44, 49341620.17it/s]
 56%|█████▌    | 2794127360/4996278331 [00:48<00:56, 39175312.35it/s]
 56%|█████▌    | 2798485504/4996278331 [00:48<00:54, 40168357.56it/s]
 56%|█████▌    | 2803924992/4996278331 [00:48<00:50, 43700552.08it/s]
 56%|█████▌    | 2809856000/4996278331 [00:48<01:11, 30600212.42it/s]
 56%|█████▋    | 2815262720/4996278331 [00:48<01:01, 35201494.72it/s]
 56%|█████▋    | 2820079616/4996278331 [00:49<00:57, 38033753.31it/s]
 57%|█████▋    | 2825617408/4996278331 [00:49<00:51, 42155751.26it/s]
 57%|█████▋    | 2832334848/4996278331 [00:49<00:44, 48439072.89it/s]
 57%|█████▋    | 2838069248/4996278331 [00:49<00:42, 50773674.62it/s]
 57%|█████▋    | 2844393472/4996278331 [00:49<00:39, 54164895.95it/s]
 57%|█████▋    | 2851209216/4996278331 [00:49<00:36, 58052639.02it/s]
 57%|█████▋    | 2857271296/4996278331 [00:49<00:36, 58301126.71it/s]
 57%|█████▋    | 2863267840/4996278331 [00:49<00:38, 56069791.47it/s]
 57%|█████▋    | 2869035008/4996278331 [00:49<00:38, 55144154.25it/s]
 58%|█████▊    | 2876080128/4996278331 [00:50<00:35, 59409601.34it/s]
 58%|█████▊    | 2882109440/4996278331 [00:50<00:35, 59085035.65it/s]
 58%|█████▊    | 2888531968/4996278331 [00:50<00:34, 60511765.12it/s]
 58%|█████▊    | 2895118336/4996278331 [00:50<00:34, 60787814.39it/s]
 58%|█████▊    | 2901245952/4996278331 [00:50<00:36, 56672157.70it/s]
 58%|█████▊    | 2907013120/4996278331 [00:50<00:37, 55808892.72it/s]
 58%|█████▊    | 2912649216/4996278331 [00:50<00:39, 52583131.06it/s]
 58%|█████▊    | 2917990400/4996278331 [00:50<00:39, 51988524.21it/s]
 59%|█████▊    | 2923692032/4996278331 [00:50<00:38, 53267557.31it/s]
 59%|█████▊    | 2929459200/4996278331 [00:51<00:37, 54509777.38it/s]
 59%|█████▊    | 2934964224/4996278331 [00:51<00:38, 53158510.59it/s]
 59%|█████▉    | 2941059072/4996278331 [00:51<00:37, 55348154.34it/s]
 59%|█████▉    | 2947645440/4996278331 [00:51<00:35, 58392210.17it/s]
 59%|█████▉    | 2955378688/4996278331 [00:51<00:31, 63909226.38it/s]
 59%|█████▉    | 2961833984/4996278331 [00:51<00:32, 63223389.10it/s]
 59%|█████▉    | 2969370624/4996278331 [00:51<00:30, 66311135.28it/s]
 60%|█████▉    | 2976022528/4996278331 [00:51<00:31, 64442244.53it/s]
 60%|█████▉    | 2982510592/4996278331 [00:51<00:34, 58513444.33it/s]
 60%|█████▉    | 2988474368/4996278331 [00:51<00:36, 54287718.15it/s]
 60%|█████▉    | 2994569216/4996278331 [00:52<00:35, 56041104.54it/s]
 60%|██████    | 3000631296/4996278331 [00:52<00:34, 57225890.21it/s]
 60%|██████    | 3008036864/4996278331 [00:52<00:32, 61950410.13it/s]
 60%|██████    | 3014328320/4996278331 [00:52<00:31, 62169754.01it/s]
 60%|██████    | 3020619776/4996278331 [00:52<00:31, 61859776.95it/s]
 61%|██████    | 3026878464/4996278331 [00:52<00:33, 59419695.03it/s]
 61%|██████    | 3032875008/4996278331 [00:52<00:34, 57495593.86it/s]
 61%|██████    | 3038707712/4996278331 [00:52<00:33, 57679184.53it/s]
 61%|██████    | 3044507648/4996278331 [00:52<00:34, 56117012.35it/s]
 61%|██████    | 3050143744/4996278331 [00:53<00:36, 53171303.17it/s]
 61%|██████    | 3055517696/4996278331 [00:53<00:37, 51773326.59it/s]
 61%|██████▏   | 3060727808/4996278331 [00:53<00:37, 51627627.30it/s]
 61%|██████▏   | 3066691584/4996278331 [00:53<00:35, 53894110.44it/s]
 61%|██████▏   | 3072131072/4996278331 [00:53<00:36, 53036144.46it/s]
 62%|██████▏   | 3077472256/4996278331 [00:53<00:37, 51718396.54it/s]
 62%|██████▏   | 3082715136/4996278331 [00:53<00:36, 51822250.84it/s]
 62%|██████▏   | 3088285696/4996278331 [00:53<00:36, 52853206.71it/s]
 62%|██████▏   | 3093594112/4996278331 [00:53<00:36, 51427708.63it/s]
 62%|██████▏   | 3098771456/4996278331 [00:54<00:41, 46156475.83it/s]
 62%|██████▏   | 3103719424/4996278331 [00:54<00:40, 47016768.10it/s]
 62%|██████▏   | 3109453824/4996278331 [00:54<00:37, 49875658.21it/s]
 62%|██████▏   | 3115581440/4996278331 [00:54<00:35, 52473748.87it/s]
 62%|██████▏   | 3121152000/4996278331 [00:54<00:35, 53123733.86it/s]
 63%|██████▎   | 3127869440/4996278331 [00:54<00:32, 57136922.69it/s]
 63%|██████▎   | 3134128128/4996278331 [00:54<00:31, 58718318.90it/s]
 63%|██████▎   | 3140386816/4996278331 [00:54<00:31, 59798985.45it/s]
 63%|██████▎   | 3147464704/4996278331 [00:54<00:29, 63026167.60it/s]
 63%|██████▎   | 3153821696/4996278331 [00:54<00:29, 62690416.35it/s]
 63%|██████▎   | 3160113152/4996278331 [00:55<00:29, 62272941.58it/s]
 63%|██████▎   | 3166371840/4996278331 [00:55<00:31, 57222227.49it/s]
 63%|██████▎   | 3172204544/4996278331 [00:55<00:33, 53749149.22it/s]
 64%|██████▎   | 3178627072/4996278331 [00:55<00:32, 56572632.94it/s]
 64%|██████▍   | 3185278976/4996278331 [00:55<00:30, 59346950.52it/s]
 64%|██████▍   | 3191308288/4996278331 [00:55<00:30, 58991774.19it/s]
 64%|██████▍   | 3198025728/4996278331 [00:55<00:29, 61266181.01it/s]
 64%|██████▍   | 3204218880/4996278331 [00:55<00:29, 59963988.35it/s]
 64%|██████▍   | 3210280960/4996278331 [00:55<00:30, 57725475.89it/s]
 64%|██████▍   | 3216769024/4996278331 [00:56<00:29, 59664652.28it/s]
 65%|██████▍   | 3222798336/4996278331 [00:56<00:29, 59313865.61it/s]
 65%|██████▍   | 3230007296/4996278331 [00:56<00:28, 62942115.80it/s]
 65%|██████▍   | 3237773312/4996278331 [00:56<00:26, 67222838.57it/s]
 65%|██████▍   | 3244556288/4996278331 [00:56<00:26, 67362883.43it/s]
 65%|██████▌   | 3251339264/4996278331 [00:56<00:28, 60323366.50it/s]
 65%|██████▌   | 3257532416/4996278331 [00:56<00:30, 56911014.17it/s]
 65%|██████▌   | 3263430656/4996278331 [00:56<00:30, 57428048.08it/s]
 65%|██████▌   | 3270049792/4996278331 [00:56<00:28, 59842910.11it/s]
 66%|██████▌   | 3276144640/4996278331 [00:57<00:28, 59508415.14it/s]
 66%|██████▌   | 3282599936/4996278331 [00:57<00:28, 60945609.73it/s]
 66%|██████▌   | 3288825856/4996278331 [00:57<00:27, 61308063.65it/s]
 66%|██████▌   | 3295019008/4996278331 [00:57<00:30, 55727743.55it/s]
 66%|██████▌   | 3300851712/4996278331 [00:57<00:30, 56380272.70it/s]
 66%|██████▌   | 3307208704/4996278331 [00:57<00:28, 58394469.58it/s]
 66%|██████▋   | 3314319360/4996278331 [00:57<00:27, 62027088.87it/s]
 66%|██████▋   | 3320610816/4996278331 [00:57<00:27, 62027368.27it/s]
 67%|██████▋   | 3326869504/4996278331 [00:57<00:27, 60796731.44it/s]
 67%|██████▋   | 3332997120/4996278331 [00:57<00:27, 59435670.80it/s]
 67%|██████▋   | 3339059200/4996278331 [00:58<00:27, 59744227.94it/s]
 67%|██████▋   | 3346825216/4996278331 [00:58<00:25, 64959348.80it/s]
 67%|██████▋   | 3353444352/4996278331 [00:58<00:25, 65314321.27it/s]
 67%|██████▋   | 3360030720/4996278331 [00:58<00:25, 65402298.74it/s]
 67%|██████▋   | 3367043072/4996278331 [00:58<00:24, 66698556.61it/s]
 68%|██████▊   | 3373727744/4996278331 [00:58<00:25, 64696386.96it/s]
 68%|██████▊   | 3380248576/4996278331 [00:58<00:26, 61693305.05it/s]
 68%|██████▊   | 3386474496/4996278331 [00:58<00:26, 61048909.60it/s]
 68%|██████▊   | 3392634880/4996278331 [00:58<00:27, 58395286.55it/s]
 68%|██████▊   | 3398533120/4996278331 [00:59<00:30, 53248646.82it/s]
 68%|██████▊   | 3405086720/4996278331 [00:59<00:28, 56508430.60it/s]
 68%|██████▊   | 3410984960/4996278331 [00:59<00:27, 57141809.38it/s]
 68%|██████▊   | 3416883200/4996278331 [00:59<00:27, 57636344.37it/s]
 69%|██████▊   | 3422715904/4996278331 [00:59<00:27, 57577684.55it/s]
 69%|██████▊   | 3429302272/4996278331 [00:59<00:26, 58430248.46it/s]
 69%|██████▉   | 3435200512/4996278331 [00:59<00:26, 57846222.48it/s]
 69%|██████▉   | 3441098752/4996278331 [00:59<00:26, 58038516.42it/s]
 69%|██████▉   | 3446931456/4996278331 [00:59<00:26, 57926905.71it/s]
 69%|██████▉   | 3452895232/4996278331 [00:59<00:26, 58350845.14it/s]
 69%|██████▉   | 3458760704/4996278331 [01:00<00:26, 57240051.07it/s]
 69%|██████▉   | 3465248768/4996278331 [01:00<00:25, 59401222.59it/s]
 70%|██████▉   | 3472588800/4996278331 [01:00<00:24, 63452062.92it/s]
 70%|██████▉   | 3478978560/4996278331 [01:00<00:24, 62058255.76it/s]
 70%|██████▉   | 3485204480/4996278331 [01:00<00:24, 61989265.03it/s]
 70%|██████▉   | 3491430400/4996278331 [01:00<00:24, 60531689.95it/s]
 70%|███████   | 3497918464/4996278331 [01:00<00:24, 61793655.91it/s]
 70%|███████   | 3505225728/4996278331 [01:00<00:22, 65032683.90it/s]
 70%|███████   | 3513122816/4996278331 [01:00<00:21, 69072445.43it/s]
 70%|███████   | 3520069632/4996278331 [01:00<00:21, 69061209.89it/s]
 71%|███████   | 3527016448/4996278331 [01:01<00:21, 68285889.07it/s]
 71%|███████   | 3533864960/4996278331 [01:01<00:22, 66004848.36it/s]
 71%|███████   | 3540516864/4996278331 [01:01<00:22, 63958630.45it/s]
 71%|███████   | 3546939392/4996278331 [01:01<00:23, 62362110.35it/s]
 71%|███████   | 3553951744/4996278331 [01:01<00:22, 64504509.82it/s]
 71%|███████▏  | 3561652224/4996278331 [01:01<00:21, 67649787.99it/s]
 71%|███████▏  | 3568467968/4996278331 [01:01<00:21, 65556946.82it/s]
 72%|███████▏  | 3575054336/4996278331 [01:01<00:22, 62686020.70it/s]
 72%|███████▏  | 3581378560/4996278331 [01:01<00:23, 60141597.07it/s]
 72%|███████▏  | 3587440640/4996278331 [01:02<00:26, 53648907.72it/s]
 72%|███████▏  | 3592945664/4996278331 [01:02<00:26, 52666196.87it/s]
 72%|███████▏  | 3599368192/4996278331 [01:02<00:25, 55726989.03it/s]
 72%|███████▏  | 3606413312/4996278331 [01:02<00:23, 59772097.22it/s]
 72%|███████▏  | 3613458432/4996278331 [01:02<00:22, 62649236.35it/s]
 72%|███████▏  | 3620110336/4996278331 [01:02<00:21, 63666860.38it/s]
 73%|███████▎  | 3626565632/4996278331 [01:02<00:22, 61450843.43it/s]
 73%|███████▎  | 3632889856/4996278331 [01:02<00:22, 61954931.00it/s]
 73%|███████▎  | 3639148544/4996278331 [01:02<00:23, 58750697.03it/s]
 73%|███████▎  | 3645112320/4996278331 [01:03<00:23, 56717729.29it/s]
 73%|███████▎  | 3650846720/4996278331 [01:03<00:23, 56675848.64it/s]
 73%|███████▎  | 3656679424/4996278331 [01:03<00:23, 57072052.91it/s]
 73%|███████▎  | 3662413824/4996278331 [01:03<00:23, 55990219.95it/s]
 73%|███████▎  | 3668049920/4996278331 [01:03<00:24, 53173928.70it/s]
 74%|███████▎  | 3673423872/4996278331 [01:03<00:25, 51243124.32it/s]
 74%|███████▎  | 3678601216/4996278331 [01:03<00:27, 48378731.62it/s]
 74%|███████▎  | 3684139008/4996278331 [01:03<00:26, 50278498.17it/s]
 74%|███████▍  | 3689775104/4996278331 [01:03<00:25, 51935569.74it/s]
 74%|███████▍  | 3695247360/4996278331 [01:04<00:24, 52698538.48it/s]
 74%|███████▍  | 3700916224/4996278331 [01:04<00:24, 53501191.50it/s]
 74%|███████▍  | 3706322944/4996278331 [01:04<00:24, 53629059.50it/s]
 74%|███████▍  | 3711729664/4996278331 [01:04<00:25, 50067824.92it/s]
 74%|███████▍  | 3716808704/4996278331 [01:04<00:27, 46464605.29it/s]
 75%|███████▍  | 3722412032/4996278331 [01:04<00:25, 49008136.06it/s]
 75%|███████▍  | 3729555456/4996278331 [01:04<00:22, 55244014.94it/s]
 75%|███████▍  | 3736928256/4996278331 [01:04<00:20, 60384402.80it/s]
 75%|███████▍  | 3743449088/4996278331 [01:04<00:20, 61706710.02it/s]
 75%|███████▌  | 3749707776/4996278331 [01:05<00:20, 61537230.00it/s]
 75%|███████▌  | 3755933696/4996278331 [01:05<00:20, 61218011.46it/s]
 75%|███████▌  | 3762094080/4996278331 [01:05<00:22, 54819255.45it/s]
 75%|███████▌  | 3767959552/4996278331 [01:05<00:21, 55835317.11it/s]
 76%|███████▌  | 3774349312/4996278331 [01:05<00:21, 58076256.63it/s]
 76%|███████▌  | 3780313088/4996278331 [01:05<00:20, 58467749.43it/s]
 76%|███████▌  | 3786670080/4996278331 [01:05<00:20, 59913713.06it/s]
 76%|███████▌  | 3793190912/4996278331 [01:05<00:19, 61407581.86it/s]
 76%|███████▌  | 3799384064/4996278331 [01:05<00:20, 57747889.29it/s]
 76%|███████▌  | 3805249536/4996278331 [01:05<00:21, 55473589.90it/s]
 76%|███████▋  | 3812229120/4996278331 [01:06<00:19, 59438371.69it/s]
 76%|███████▋  | 3819667456/4996278331 [01:06<00:18, 63649124.77it/s]
 77%|███████▋  | 3826122752/4996278331 [01:06<00:18, 63736713.68it/s]
 77%|███████▋  | 3832971264/4996278331 [01:06<00:17, 65073427.89it/s]
 77%|███████▋  | 3839524864/4996278331 [01:06<00:18, 63926101.61it/s]
 77%|███████▋  | 3845980160/4996278331 [01:06<00:18, 60954311.90it/s]
 77%|███████▋  | 3852140544/4996278331 [01:06<00:19, 57678152.75it/s]
 77%|███████▋  | 3858104320/4996278331 [01:06<00:19, 58176630.85it/s]
 77%|███████▋  | 3864231936/4996278331 [01:06<00:19, 58988738.33it/s]
 77%|███████▋  | 3870195712/4996278331 [01:07<00:19, 57158847.88it/s]
 78%|███████▊  | 3876093952/4996278331 [01:07<00:19, 57666209.36it/s]
 78%|███████▊  | 3882024960/4996278331 [01:07<00:19, 58062388.27it/s]
 78%|███████▊  | 3887857664/4996278331 [01:07<00:20, 54831434.48it/s]
 78%|███████▊  | 3893395456/4996278331 [01:07<00:20, 54397586.06it/s]
 78%|███████▊  | 3899719680/4996278331 [01:07<00:19, 56861162.52it/s]
 78%|███████▊  | 3906469888/4996278331 [01:07<00:18, 59880216.39it/s]
 78%|███████▊  | 3912925184/4996278331 [01:07<00:17, 61243496.97it/s]
 78%|███████▊  | 3919085568/4996278331 [01:07<00:17, 60369920.58it/s]
 79%|███████▊  | 3925147648/4996278331 [01:08<00:18, 57168255.70it/s]
 79%|███████▊  | 3931602944/4996278331 [01:08<00:17, 59231814.38it/s]
 79%|███████▉  | 3937599488/4996278331 [01:08<00:18, 58707634.61it/s]
 79%|███████▉  | 3943530496/4996278331 [01:08<00:19, 55285880.73it/s]
 79%|███████▉  | 3949133824/4996278331 [01:08<00:19, 54167193.56it/s]
 79%|███████▉  | 3954933760/4996278331 [01:08<00:18, 55214388.99it/s]
 79%|███████▉  | 3960733696/4996278331 [01:08<00:18, 55993345.88it/s]
 79%|███████▉  | 3966369792/4996278331 [01:08<00:18, 56085395.16it/s]
 79%|███████▉  | 3972005888/4996278331 [01:08<00:18, 55159722.26it/s]
 80%|███████▉  | 3977543680/4996278331 [01:08<00:19, 51076620.22it/s]
 80%|███████▉  | 3982721024/4996278331 [01:09<00:20, 50473255.58it/s]
 80%|███████▉  | 3988881408/4996278331 [01:09<00:18, 53573545.64it/s]
 80%|███████▉  | 3994484736/4996278331 [01:09<00:18, 54230685.21it/s]
 80%|████████  | 4000546816/4996278331 [01:09<00:17, 55911030.21it/s]
 80%|████████  | 4006182912/4996278331 [01:09<00:17, 56038253.18it/s]
 80%|████████  | 4011819008/4996278331 [01:09<00:18, 53934863.55it/s]
 80%|████████  | 4017258496/4996278331 [01:09<00:18, 52707556.66it/s]
 81%|████████  | 4022566912/4996278331 [01:09<00:18, 52139797.18it/s]
 81%|████████  | 4028071936/4996278331 [01:09<00:18, 52946544.01it/s]
 81%|████████  | 4033413120/4996278331 [01:10<00:18, 51784828.92it/s]
 81%|████████  | 4038983680/4996278331 [01:10<00:18, 52830555.57it/s]
 81%|████████  | 4044947456/4996278331 [01:10<00:17, 54776007.05it/s]
 81%|████████  | 4050452480/4996278331 [01:10<00:17, 53836566.39it/s]
 81%|████████  | 4055859200/4996278331 [01:10<00:17, 53338070.80it/s]
 81%|████████▏ | 4061429760/4996278331 [01:10<00:17, 54002010.05it/s]
 81%|████████▏ | 4067557376/4996278331 [01:10<00:16, 56084047.63it/s]
 82%|████████▏ | 4074471424/4996278331 [01:10<00:15, 59916182.64it/s]
 82%|████████▏ | 4080500736/4996278331 [01:10<00:15, 59936174.47it/s]
 82%|████████▏ | 4086530048/4996278331 [01:10<00:15, 57640866.64it/s]
 82%|████████▏ | 4092329984/4996278331 [01:11<00:16, 53805978.67it/s]
 82%|████████▏ | 4097900544/4996278331 [01:11<00:16, 54239635.20it/s]
 82%|████████▏ | 4103372800/4996278331 [01:11<00:16, 53600162.28it/s]
 82%|████████▏ | 4109336576/4996278331 [01:11<00:16, 55282339.97it/s]
 82%|████████▏ | 4116185088/4996278331 [01:11<00:14, 59087139.88it/s]
 83%|████████▎ | 4122148864/4996278331 [01:11<00:14, 58391837.79it/s]
 83%|████████▎ | 4128407552/4996278331 [01:11<00:14, 59607166.98it/s]
 83%|████████▎ | 4134404096/4996278331 [01:11<00:15, 56883666.70it/s]
 83%|████████▎ | 4140138496/4996278331 [01:11<00:15, 54168032.44it/s]
 83%|████████▎ | 4146003968/4996278331 [01:12<00:15, 55354518.24it/s]
 83%|████████▎ | 4152754176/4996278331 [01:12<00:14, 58804355.42it/s]
 83%|████████▎ | 4160192512/4996278331 [01:12<00:13, 63296849.67it/s]
 83%|████████▎ | 4166746112/4996278331 [01:12<00:12, 63875535.12it/s]
 84%|████████▎ | 4173168640/4996278331 [01:12<00:13, 60139218.44it/s]
 84%|████████▎ | 4179263488/4996278331 [01:12<00:15, 53975099.85it/s]
 84%|████████▍ | 4184899584/4996278331 [01:12<00:14, 54499800.08it/s]
 84%|████████▍ | 4190470144/4996278331 [01:12<00:14, 53878751.36it/s]
 84%|████████▍ | 4195942400/4996278331 [01:12<00:15, 51819630.47it/s]
 84%|████████▍ | 4202463232/4996278331 [01:12<00:14, 55509952.00it/s]
 84%|████████▍ | 4208787456/4996278331 [01:13<00:13, 57637411.39it/s]
 84%|████████▍ | 4214816768/4996278331 [01:13<00:13, 58275270.59it/s]
 84%|████████▍ | 4220715008/4996278331 [01:13<00:13, 58300627.15it/s]
 85%|████████▍ | 4227268608/4996278331 [01:13<00:12, 60381130.21it/s]
 85%|████████▍ | 4233363456/4996278331 [01:13<00:12, 59466320.27it/s]
 85%|████████▍ | 4239491072/4996278331 [01:13<00:12, 59940217.70it/s]
 85%|████████▍ | 4245520384/4996278331 [01:13<00:12, 58297932.26it/s]
 85%|████████▌ | 4251385856/4996278331 [01:13<00:13, 57061081.60it/s]
 85%|████████▌ | 4257120256/4996278331 [01:13<00:13, 55408233.51it/s]
 85%|████████▌ | 4262690816/4996278331 [01:14<00:13, 55221398.65it/s]
 85%|████████▌ | 4268228608/4996278331 [01:14<00:14, 51417360.68it/s]
 86%|████████▌ | 4273438720/4996278331 [01:14<00:14, 48507762.32it/s]
 86%|████████▌ | 4278353920/4996278331 [01:14<00:17, 42049088.42it/s]
 86%|████████▌ | 4282744832/4996278331 [01:14<00:17, 39902197.52it/s]
 86%|████████▌ | 4287266816/4996278331 [01:14<00:17, 41107341.39it/s]
 86%|████████▌ | 4292706304/4996278331 [01:14<00:15, 44564070.38it/s]
 86%|████████▌ | 4298670080/4996278331 [01:14<00:14, 48659594.30it/s]
 86%|████████▌ | 4304273408/4996278331 [01:14<00:13, 50650832.83it/s]
 86%|████████▋ | 4310466560/4996278331 [01:15<00:12, 53815455.87it/s]
 86%|████████▋ | 4317020160/4996278331 [01:15<00:11, 57139349.77it/s]
 87%|████████▋ | 4322983936/4996278331 [01:15<00:11, 57819671.32it/s]
 87%|████████▋ | 4329930752/4996278331 [01:15<00:10, 61223414.09it/s]
 87%|████████▋ | 4336123904/4996278331 [01:15<00:11, 57767636.72it/s]
 87%|████████▋ | 4341989376/4996278331 [01:15<00:11, 54761573.66it/s]
 87%|████████▋ | 4347559936/4996278331 [01:15<00:12, 52390964.72it/s]
 87%|████████▋ | 4354179072/4996278331 [01:15<00:11, 56148667.51it/s]
 87%|████████▋ | 4361682944/4996278331 [01:15<00:10, 61451248.44it/s]
 87%|████████▋ | 4367941632/4996278331 [01:16<00:10, 61348387.66it/s]
 88%|████████▊ | 4375379968/4996278331 [01:16<00:09, 65095679.48it/s]
 88%|████████▊ | 4381966336/4996278331 [01:16<00:09, 64125642.37it/s]
 88%|████████▊ | 4388421632/4996278331 [01:16<00:09, 61503456.40it/s]
 88%|████████▊ | 4394647552/4996278331 [01:16<00:09, 60439875.17it/s]
 88%|████████▊ | 4400742400/4996278331 [01:16<00:09, 59825856.50it/s]
 88%|████████▊ | 4407623680/4996278331 [01:16<00:09, 62334075.86it/s]
 88%|████████▊ | 4413915136/4996278331 [01:16<00:09, 61237944.38it/s]
 88%|████████▊ | 4420567040/4996278331 [01:16<00:09, 62702104.87it/s]
 89%|████████▊ | 4426858496/4996278331 [01:16<00:09, 60036738.19it/s]
 89%|████████▊ | 4432920576/4996278331 [01:17<00:09, 59668606.31it/s]
 89%|████████▉ | 4438917120/4996278331 [01:17<00:09, 57975768.47it/s]
 89%|████████▉ | 4444880896/4996278331 [01:17<00:09, 58368816.22it/s]
 89%|████████▉ | 4450746368/4996278331 [01:17<00:09, 57452222.12it/s]
 89%|████████▉ | 4456873984/4996278331 [01:17<00:09, 58505720.52it/s]
 89%|████████▉ | 4464017408/4996278331 [01:17<00:08, 62215394.68it/s]
 89%|████████▉ | 4470276096/4996278331 [01:17<00:08, 61349890.42it/s]
 90%|████████▉ | 4476436480/4996278331 [01:17<00:08, 58067679.48it/s]
 90%|████████▉ | 4482301952/4996278331 [01:17<00:09, 55220592.54it/s]
 90%|████████▉ | 4488232960/4996278331 [01:18<00:09, 56313479.98it/s]
 90%|████████▉ | 4494458880/4996278331 [01:18<00:08, 57994827.09it/s]
 90%|█████████ | 4500324352/4996278331 [01:18<00:08, 58064195.30it/s]
 90%|█████████ | 4506189824/4996278331 [01:18<00:09, 53068524.26it/s]
 90%|█████████ | 4511727616/4996278331 [01:18<00:09, 53539325.47it/s]
 90%|█████████ | 4517724160/4996278331 [01:18<00:08, 55133848.78it/s]
 91%|█████████ | 4523327488/4996278331 [01:18<00:09, 52002334.87it/s]
 91%|█████████ | 4528603136/4996278331 [01:18<00:09, 51846686.16it/s]
 91%|█████████ | 4533977088/4996278331 [01:18<00:08, 52333031.52it/s]
 91%|█████████ | 4539252736/4996278331 [01:19<00:08, 52298459.51it/s]
 91%|█████████ | 4544757760/4996278331 [01:19<00:08, 52988661.60it/s]
 91%|█████████ | 4550098944/4996278331 [01:19<00:08, 52921494.56it/s]
 91%|█████████ | 4555407360/4996278331 [01:19<00:09, 45734639.85it/s]
 91%|█████████▏| 4560977920/4996278331 [01:19<00:08, 48374990.00it/s]
 91%|█████████▏| 4567400448/4996278331 [01:19<00:08, 52716574.85it/s]
 92%|█████████▏| 4573593600/4996278331 [01:19<00:07, 55314795.61it/s]
 92%|█████████▏| 4579786752/4996278331 [01:19<00:07, 57152463.99it/s]
 92%|█████████▏| 4585619456/4996278331 [01:19<00:07, 57003377.50it/s]
 92%|█████████▏| 4591386624/4996278331 [01:20<00:07, 53526388.24it/s]
 92%|█████████▏| 4596826112/4996278331 [01:20<00:07, 53320616.63it/s]
 92%|█████████▏| 4603150336/4996278331 [01:20<00:07, 56122851.52it/s]
 92%|█████████▏| 4610228224/4996278331 [01:20<00:06, 60344764.58it/s]
 92%|█████████▏| 4617666560/4996278331 [01:20<00:05, 64418033.60it/s]
 93%|█████████▎| 4624187392/4996278331 [01:20<00:06, 60188264.56it/s]
 93%|█████████▎| 4630315008/4996278331 [01:20<00:06, 59398388.54it/s]
 93%|█████████▎| 4636442624/4996278331 [01:20<00:06, 59260616.71it/s]
 93%|█████████▎| 4642439168/4996278331 [01:20<00:06, 58280262.86it/s]
 93%|█████████▎| 4648960000/4996278331 [01:20<00:05, 60180872.16it/s]
 93%|█████████▎| 4656398336/4996278331 [01:21<00:05, 64279551.21it/s]
 93%|█████████▎| 4664590336/4996278331 [01:21<00:04, 69372815.56it/s]
 94%|█████████▎| 4671569920/4996278331 [01:21<00:04, 67606773.84it/s]
 94%|█████████▎| 4678385664/4996278331 [01:21<00:04, 64666249.91it/s]
 94%|█████████▍| 4684906496/4996278331 [01:21<00:04, 62685130.91it/s]
 94%|█████████▍| 4692115456/4996278331 [01:21<00:04, 65313454.43it/s]
 94%|█████████▍| 4699979776/4996278331 [01:21<00:04, 69117056.99it/s]
 94%|█████████▍| 4707385344/4996278331 [01:21<00:04, 70504794.99it/s]
 94%|█████████▍| 4715118592/4996278331 [01:21<00:03, 72448571.99it/s]
 95%|█████████▍| 4722425856/4996278331 [01:22<00:04, 67185885.16it/s]
 95%|█████████▍| 4729241600/4996278331 [01:22<00:04, 64061893.57it/s]
 95%|█████████▍| 4735762432/4996278331 [01:22<00:04, 61481636.46it/s]
 95%|█████████▍| 4742938624/4996278331 [01:22<00:03, 64214484.81it/s]
 95%|█████████▌| 4749819904/4996278331 [01:22<00:03, 65389401.66it/s]
 95%|█████████▌| 4756439040/4996278331 [01:22<00:03, 65054505.71it/s]
 95%|█████████▌| 4763877376/4996278331 [01:22<00:03, 67724038.24it/s]
 95%|█████████▌| 4770693120/4996278331 [01:22<00:03, 62993811.98it/s]
 96%|█████████▌| 4777082880/4996278331 [01:22<00:03, 56331033.57it/s]
 96%|█████████▌| 4782915584/4996278331 [01:23<00:03, 55383504.97it/s]
 96%|█████████▌| 4788715520/4996278331 [01:23<00:03, 56078741.33it/s]
 96%|█████████▌| 4795662336/4996278331 [01:23<00:03, 59724665.68it/s]
 96%|█████████▌| 4801757184/4996278331 [01:23<00:03, 59308091.44it/s]
 96%|█████████▌| 4807786496/4996278331 [01:23<00:03, 59571480.07it/s]
 96%|█████████▋| 4813815808/4996278331 [01:23<00:03, 59543985.85it/s]
 96%|█████████▋| 4819812352/4996278331 [01:23<00:03, 58688704.38it/s]
 97%|█████████▋| 4826529792/4996278331 [01:23<00:02, 61098358.45it/s]
 97%|█████████▋| 4834000896/4996278331 [01:23<00:02, 65083092.25it/s]
 97%|█████████▋| 4840751104/4996278331 [01:23<00:02, 65790670.44it/s]
 97%|█████████▋| 4847370240/4996278331 [01:24<00:02, 61695741.16it/s]
 97%|█████████▋| 4853628928/4996278331 [01:24<00:02, 60504746.09it/s]
 97%|█████████▋| 4860116992/4996278331 [01:24<00:02, 61664291.26it/s]
 97%|█████████▋| 4866342912/4996278331 [01:24<00:02, 61038972.96it/s]
 98%|█████████▊| 4872994816/4996278331 [01:24<00:01, 62597484.02it/s]
 98%|█████████▊| 4880531456/4996278331 [01:24<00:01, 66283721.95it/s]
 98%|█████████▊| 4887216128/4996278331 [01:24<00:01, 65306613.82it/s]
 98%|█████████▊| 4893900800/4996278331 [01:24<00:01, 65727331.56it/s]
 98%|█████████▊| 4900519936/4996278331 [01:24<00:01, 63601089.96it/s]
 98%|█████████▊| 4906909696/4996278331 [01:25<00:01, 59782885.42it/s]
 98%|█████████▊| 4912971776/4996278331 [01:25<00:01, 59497886.11it/s]
 98%|█████████▊| 4920541184/4996278331 [01:25<00:01, 63997192.90it/s]
 99%|█████████▊| 4926996480/4996278331 [01:25<00:01, 63520697.14it/s]
 99%|█████████▊| 4933386240/4996278331 [01:25<00:01, 62886496.41it/s]
 99%|█████████▉| 4939710464/4996278331 [01:25<00:00, 60848067.11it/s]
 99%|█████████▉| 4945838080/4996278331 [01:25<00:00, 59425385.06it/s]
 99%|█████████▉| 4951801856/4996278331 [01:25<00:00, 58494168.80it/s]
 99%|█████████▉| 4957667328/4996278331 [01:25<00:00, 56779601.80it/s]
 99%|█████████▉| 4963368960/4996278331 [01:26<00:00, 51995108.79it/s]
 99%|█████████▉| 4968644608/4996278331 [01:26<00:00, 50278629.09it/s]
100%|█████████▉| 4973723648/4996278331 [01:26<00:00, 49210039.91it/s]
100%|█████████▉| 4978704384/4996278331 [01:26<00:00, 43142245.47it/s]
100%|█████████▉| 4983357440/4996278331 [01:26<00:00, 43994542.93it/s]
100%|█████████▉| 4988534784/4996278331 [01:26<00:00, 46033596.23it/s]
100%|█████████▉| 4995121152/4996278331 [01:26<00:00, 51419521.15it/s]
100%|██████████| 4996278331/4996278331 [01:26<00:00, 57617532.77it/s]
Extracting datasets/food101/food-101.tar.gz to datasets/food101

Load Model Checkpoint

Having chosen a model checkpoint to load, we will now create a PyTorch Module using the pre-trained weights.

from torchvision.models import resnet50

# Initialize the ResNet-50 model
model = resnet50()
num_feats = model.fc.in_features

# Load the checkpoint
checkpoint = torch.hub.load_state_dict_from_url(url=checkpoint_url, map_location=device)

# Remove 'backbone.' prefix from keys in state_dict
backbone_state_dict = {
    k.replace("backbone.", ""): v
    for k, v in checkpoint["state_dict"].items()
    if k.startswith("backbone.")
}

# Load the state_dict into the model
model.load_state_dict(backbone_state_dict, strict=False)

# Add a classification head
model.fc = nn.Linear(num_feats, num_classes)

Create a FineTuning Lightning Module

Now let’s create a PyTorch Lightning Module to finetune our pre-trained model.

from torch.optim import AdamW
from torchmetrics import Accuracy


class FinetuningModel(pl.LightningModule):
    def __init__(self, model, num_classes: int, lr: float):
        super().__init__()
        self.save_hyperparameters()
        self.model = model
        self.train_accuracy = Accuracy(task="multiclass", num_classes=num_classes)
        self.val_accuracy = Accuracy(task="multiclass", num_classes=num_classes)
        self.loss_fn = nn.CrossEntropyLoss()

    def forward(self, x):
        return self.model(x)

    def training_step(self, batch, batch_idx):
        x, y = batch
        logits = self(x)
        loss = self.loss_fn(logits, y)
        self.log("train_loss", loss)
        preds = torch.argmax(logits, dim=1)
        acc = self.train_accuracy(preds, y)
        self.log("train_accuracy", acc)
        return loss

    def validation_step(self, batch, batch_idx):
        x, y = batch
        logits = self(x)
        loss = self.loss_fn(logits, y)
        self.log("val_loss", loss)
        preds = torch.argmax(logits, dim=1)
        acc = self.val_accuracy(preds, y)
        self.log("val_accuracy", acc)
        return loss

    def configure_optimizers(self):
        optimizer = AdamW(self.parameters(), lr=self.hparams.lr)
        return optimizer

Training

Let’s instantiate our Lightning Module and train it using the Lightning Trainer.

finetuning_model = FinetuningModel(model, lr=learning_rate, num_classes=num_classes)

trainer = pl.Trainer(
    max_epochs=num_train_epochs, devices=1, logger=tb_logger, accelerator=device
)

trainer.fit(
    model=finetuning_model,
    train_dataloaders=train_dataloader,
    val_dataloaders=val_dataloader,
)
/datasets/actions-runner/core_gpu_runner_03/_work/lightly-core/lightly-core/.venv/lib/python3.10/site-packages/pytorch_lightning/utilities/parsing.py:269: UserWarning: Attribute 'model' is an instance of `nn.Module` and is already saved during checkpointing. It is recommended to ignore them using `self.save_hyperparameters(ignore=['model'])`.
  rank_zero_warn(
/datasets/actions-runner/core_gpu_runner_03/_work/lightly-core/lightly-core/.venv/lib/python3.10/site-packages/lightning_fabric/plugins/environments/slurm.py:165: PossibleUserWarning: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python3 /datasets/actions-runner/core_gpu_runner_03/_work/l ...
  rank_zero_warn(

Sanity Checking: 0it [00:00, ?it/s]/datasets/actions-runner/core_gpu_runner_03/_work/lightly-core/lightly-core/.venv/lib/python3.10/site-packages/pytorch_lightning/trainer/connectors/data_connector.py:488: PossibleUserWarning: Your `val_dataloader`'s sampler has shuffling enabled, it is strongly recommended that you turn shuffling off for val/test/predict dataloaders.
  rank_zero_warn(

Sanity Checking:   0%|          | 0/2 [00:00<?, ?it/s]
Sanity Checking DataLoader 0:   0%|          | 0/2 [00:00<?, ?it/s]
Sanity Checking DataLoader 0:  50%|█████     | 1/2 [00:01<00:01,  1.90s/it]
Sanity Checking DataLoader 0: 100%|██████████| 2/2 [00:01<00:00,  1.05it/s]


Training: 0it [00:00, ?it/s]
Training:   0%|          | 0/33666 [00:00<?, ?it/s]
Epoch 0:   0%|          | 0/33666 [00:00<?, ?it/s]
Epoch 0:   0%|          | 1/33666 [00:00<7:05:23,  1.32it/s]
Epoch 0:   0%|          | 1/33666 [00:00<7:05:56,  1.32it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 2/33666 [00:00<3:45:28,  2.49it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 2/33666 [00:00<3:45:35,  2.49it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 3/33666 [00:00<2:36:49,  3.58it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 3/33666 [00:00<2:36:53,  3.58it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 4/33666 [00:00<2:02:38,  4.57it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 4/33666 [00:00<2:02:41,  4.57it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 5/33666 [00:00<1:42:06,  5.49it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 5/33666 [00:00<1:42:08,  5.49it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 6/33666 [00:00<1:28:54,  6.31it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 6/33666 [00:00<1:28:57,  6.31it/s, loss=4.57, v_num=0]
Epoch 0:   0%|          | 7/33666 [00:00<1:19:17,  7.07it/s, loss=4.57, v_num=0]
Epoch 0:   0%|          | 7/33666 [00:00<1:19:19,  7.07it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 8/33666 [00:01<1:11:47,  7.81it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 8/33666 [00:01<1:11:48,  7.81it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 9/33666 [00:01<1:06:00,  8.50it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 9/33666 [00:01<1:06:02,  8.49it/s, loss=4.58, v_num=0]
Epoch 0:   0%|          | 10/33666 [00:01<1:01:41,  9.09it/s, loss=4.58, v_num=0]
Epoch 0:   0%|          | 10/33666 [00:01<1:01:42,  9.09it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 11/33666 [00:01<57:50,  9.70it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 11/33666 [00:01<57:51,  9.69it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 12/33666 [00:01<54:40, 10.26it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 12/33666 [00:01<54:41, 10.26it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 13/33666 [00:01<52:01, 10.78it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 13/33666 [00:01<52:02, 10.78it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 14/33666 [00:01<49:38, 11.30it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 14/33666 [00:01<49:39, 11.29it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 15/33666 [00:01<47:40, 11.76it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 15/33666 [00:01<47:41, 11.76it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 16/33666 [00:01<45:53, 12.22it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 16/33666 [00:01<45:54, 12.22it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 17/33666 [00:01<44:14, 12.68it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 17/33666 [00:01<44:14, 12.67it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 18/33666 [00:01<42:59, 13.04it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 18/33666 [00:01<43:00, 13.04it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 19/33666 [00:01<41:59, 13.36it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 19/33666 [00:01<41:59, 13.35it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 20/33666 [00:01<40:56, 13.70it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 20/33666 [00:01<40:57, 13.69it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 21/33666 [00:01<40:07, 13.98it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 21/33666 [00:01<40:07, 13.97it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 22/33666 [00:01<39:15, 14.28it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 22/33666 [00:01<39:15, 14.28it/s, loss=4.75, v_num=0]
Epoch 0:   0%|          | 23/33666 [00:01<38:34, 14.53it/s, loss=4.75, v_num=0]
Epoch 0:   0%|          | 23/33666 [00:01<38:35, 14.53it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 24/33666 [00:01<37:51, 14.81it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 24/33666 [00:01<37:51, 14.81it/s, loss=4.75, v_num=0]
Epoch 0:   0%|          | 25/33666 [00:01<37:11, 15.07it/s, loss=4.75, v_num=0]
Epoch 0:   0%|          | 25/33666 [00:01<37:12, 15.07it/s, loss=4.74, v_num=0]
Epoch 0:   0%|          | 26/33666 [00:01<36:35, 15.32it/s, loss=4.74, v_num=0]
Epoch 0:   0%|          | 26/33666 [00:01<36:35, 15.32it/s, loss=4.72, v_num=0]
Epoch 0:   0%|          | 27/33666 [00:01<36:06, 15.53it/s, loss=4.72, v_num=0]
Epoch 0:   0%|          | 27/33666 [00:01<36:06, 15.53it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 28/33666 [00:01<35:34, 15.76it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 28/33666 [00:01<35:34, 15.76it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 29/33666 [00:01<35:09, 15.95it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 29/33666 [00:01<35:09, 15.94it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 30/33666 [00:01<34:41, 16.16it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 30/33666 [00:01<34:41, 16.16it/s, loss=4.74, v_num=0]
Epoch 0:   0%|          | 31/33666 [00:01<34:19, 16.33it/s, loss=4.74, v_num=0]
Epoch 0:   0%|          | 31/33666 [00:01<34:20, 16.33it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 32/33666 [00:01<33:55, 16.53it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 32/33666 [00:01<33:55, 16.52it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 33/33666 [00:01<33:33, 16.71it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 33/33666 [00:01<33:33, 16.70it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 34/33666 [00:02<33:12, 16.88it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 34/33666 [00:02<33:12, 16.88it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 35/33666 [00:02<32:51, 17.06it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 35/33666 [00:02<32:52, 17.05it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 36/33666 [00:02<32:36, 17.19it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 36/33666 [00:02<32:36, 17.19it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 37/33666 [00:02<32:17, 17.35it/s, loss=4.73, v_num=0]
Epoch 0:   0%|          | 37/33666 [00:02<32:18, 17.35it/s, loss=4.76, v_num=0]
Epoch 0:   0%|          | 38/33666 [00:02<32:04, 17.48it/s, loss=4.76, v_num=0]
Epoch 0:   0%|          | 38/33666 [00:02<32:04, 17.47it/s, loss=4.76, v_num=0]
Epoch 0:   0%|          | 39/33666 [00:02<31:47, 17.63it/s, loss=4.76, v_num=0]
Epoch 0:   0%|          | 39/33666 [00:02<31:47, 17.63it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 40/33666 [00:02<31:35, 17.74it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 40/33666 [00:02<31:35, 17.74it/s, loss=4.81, v_num=0]
Epoch 0:   0%|          | 41/33666 [00:02<31:18, 17.90it/s, loss=4.81, v_num=0]
Epoch 0:   0%|          | 41/33666 [00:02<31:18, 17.90it/s, loss=4.79, v_num=0]
Epoch 0:   0%|          | 42/33666 [00:02<31:04, 18.04it/s, loss=4.79, v_num=0]
Epoch 0:   0%|          | 42/33666 [00:02<31:04, 18.03it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 43/33666 [00:02<30:53, 18.14it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 43/33666 [00:02<30:53, 18.14it/s, loss=4.84, v_num=0]
Epoch 0:   0%|          | 44/33666 [00:02<30:40, 18.27it/s, loss=4.84, v_num=0]
Epoch 0:   0%|          | 44/33666 [00:02<30:40, 18.27it/s, loss=4.83, v_num=0]
Epoch 0:   0%|          | 45/33666 [00:02<30:30, 18.36it/s, loss=4.83, v_num=0]
Epoch 0:   0%|          | 45/33666 [00:02<30:31, 18.36it/s, loss=4.83, v_num=0]
Epoch 0:   0%|          | 46/33666 [00:02<30:18, 18.48it/s, loss=4.83, v_num=0]
Epoch 0:   0%|          | 46/33666 [00:02<30:19, 18.48it/s, loss=4.89, v_num=0]
Epoch 0:   0%|          | 47/33666 [00:02<30:10, 18.57it/s, loss=4.89, v_num=0]
Epoch 0:   0%|          | 47/33666 [00:02<30:10, 18.57it/s, loss=4.89, v_num=0]
Epoch 0:   0%|          | 48/33666 [00:02<29:58, 18.69it/s, loss=4.89, v_num=0]
Epoch 0:   0%|          | 48/33666 [00:02<29:59, 18.69it/s, loss=4.9, v_num=0]
Epoch 0:   0%|          | 49/33666 [00:02<29:46, 18.82it/s, loss=4.9, v_num=0]
Epoch 0:   0%|          | 49/33666 [00:02<29:46, 18.81it/s, loss=4.92, v_num=0]
Epoch 0:   0%|          | 50/33666 [00:02<29:39, 18.89it/s, loss=4.92, v_num=0]
Epoch 0:   0%|          | 50/33666 [00:02<29:39, 18.89it/s, loss=4.93, v_num=0]
Epoch 0:   0%|          | 51/33666 [00:02<29:31, 18.97it/s, loss=4.93, v_num=0]
Epoch 0:   0%|          | 51/33666 [00:02<29:32, 18.97it/s, loss=4.95, v_num=0]
Epoch 0:   0%|          | 52/33666 [00:02<29:24, 19.05it/s, loss=4.95, v_num=0]
Epoch 0:   0%|          | 52/33666 [00:02<29:24, 19.05it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 53/33666 [00:02<29:16, 19.13it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 53/33666 [00:02<29:16, 19.13it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 54/33666 [00:02<29:07, 19.23it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 54/33666 [00:02<29:08, 19.23it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 55/33666 [00:02<29:01, 19.30it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 55/33666 [00:02<29:02, 19.29it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 56/33666 [00:02<28:54, 19.37it/s, loss=4.98, v_num=0]
Epoch 0:   0%|          | 56/33666 [00:02<28:55, 19.37it/s, loss=4.94, v_num=0]
Epoch 0:   0%|          | 57/33666 [00:02<28:50, 19.42it/s, loss=4.94, v_num=0]
Epoch 0:   0%|          | 57/33666 [00:02<28:50, 19.42it/s, loss=4.93, v_num=0]
Epoch 0:   0%|          | 58/33666 [00:02<28:43, 19.50it/s, loss=4.93, v_num=0]
Epoch 0:   0%|          | 58/33666 [00:02<28:44, 19.49it/s, loss=4.93, v_num=0]
Epoch 0:   0%|          | 59/33666 [00:03<28:36, 19.58it/s, loss=4.93, v_num=0]
Epoch 0:   0%|          | 59/33666 [00:03<28:36, 19.58it/s, loss=4.9, v_num=0]
Epoch 0:   0%|          | 60/33666 [00:03<28:30, 19.64it/s, loss=4.9, v_num=0]
Epoch 0:   0%|          | 60/33666 [00:03<28:31, 19.64it/s, loss=4.86, v_num=0]
Epoch 0:   0%|          | 61/33666 [00:03<28:23, 19.73it/s, loss=4.86, v_num=0]
Epoch 0:   0%|          | 61/33666 [00:03<28:23, 19.72it/s, loss=4.84, v_num=0]
Epoch 0:   0%|          | 62/33666 [00:03<28:17, 19.80it/s, loss=4.84, v_num=0]
Epoch 0:   0%|          | 62/33666 [00:03<28:17, 19.79it/s, loss=4.81, v_num=0]
Epoch 0:   0%|          | 63/33666 [00:03<28:13, 19.85it/s, loss=4.81, v_num=0]
Epoch 0:   0%|          | 63/33666 [00:03<28:13, 19.84it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 64/33666 [00:03<28:06, 19.92it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 64/33666 [00:03<28:06, 19.92it/s, loss=4.82, v_num=0]
Epoch 0:   0%|          | 65/33666 [00:03<28:02, 19.97it/s, loss=4.82, v_num=0]
Epoch 0:   0%|          | 65/33666 [00:03<28:02, 19.97it/s, loss=4.83, v_num=0]
Epoch 0:   0%|          | 66/33666 [00:03<27:57, 20.03it/s, loss=4.83, v_num=0]
Epoch 0:   0%|          | 66/33666 [00:03<27:57, 20.03it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 67/33666 [00:03<27:50, 20.11it/s, loss=4.8, v_num=0]
Epoch 0:   0%|          | 67/33666 [00:03<27:50, 20.11it/s, loss=4.78, v_num=0]
Epoch 0:   0%|          | 68/33666 [00:03<27:43, 20.20it/s, loss=4.78, v_num=0]
Epoch 0:   0%|          | 68/33666 [00:03<27:43, 20.19it/s, loss=4.75, v_num=0]
Epoch 0:   0%|          | 69/33666 [00:03<27:36, 20.29it/s, loss=4.75, v_num=0]
Epoch 0:   0%|          | 69/33666 [00:03<27:36, 20.29it/s, loss=4.74, v_num=0]
Epoch 0:   0%|          | 70/33666 [00:03<27:29, 20.37it/s, loss=4.74, v_num=0]
Epoch 0:   0%|          | 70/33666 [00:03<27:29, 20.37it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 71/33666 [00:03<27:23, 20.44it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 71/33666 [00:03<27:23, 20.44it/s, loss=4.72, v_num=0]
Epoch 0:   0%|          | 72/33666 [00:03<27:16, 20.52it/s, loss=4.72, v_num=0]
Epoch 0:   0%|          | 72/33666 [00:03<27:16, 20.52it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 73/33666 [00:03<27:14, 20.56it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 73/33666 [00:03<27:14, 20.55it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 74/33666 [00:03<27:09, 20.61it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 74/33666 [00:03<27:10, 20.61it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 75/33666 [00:03<27:07, 20.64it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 75/33666 [00:03<27:07, 20.64it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 76/33666 [00:03<27:04, 20.68it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 76/33666 [00:03<27:04, 20.68it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 77/33666 [00:03<26:59, 20.74it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 77/33666 [00:03<26:59, 20.74it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 78/33666 [00:03<26:58, 20.76it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 78/33666 [00:03<26:58, 20.75it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 79/33666 [00:03<26:52, 20.84it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 79/33666 [00:03<26:52, 20.83it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 80/33666 [00:03<26:47, 20.89it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 80/33666 [00:03<26:47, 20.89it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 81/33666 [00:03<26:43, 20.95it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 81/33666 [00:03<26:43, 20.95it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 82/33666 [00:03<26:38, 21.01it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 82/33666 [00:03<26:38, 21.01it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 83/33666 [00:03<26:32, 21.08it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 83/33666 [00:03<26:33, 21.08it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 84/33666 [00:03<26:27, 21.15it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 84/33666 [00:03<26:27, 21.15it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 85/33666 [00:04<26:22, 21.21it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 85/33666 [00:04<26:23, 21.21it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 86/33666 [00:04<26:18, 21.27it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 86/33666 [00:04<26:18, 21.27it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 87/33666 [00:04<26:14, 21.33it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 87/33666 [00:04<26:14, 21.33it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 88/33666 [00:04<26:09, 21.39it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 88/33666 [00:04<26:09, 21.39it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 89/33666 [00:04<26:07, 21.42it/s, loss=4.59, v_num=0]
Epoch 0:   0%|          | 89/33666 [00:04<26:07, 21.42it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 90/33666 [00:04<26:01, 21.50it/s, loss=4.6, v_num=0]
Epoch 0:   0%|          | 90/33666 [00:04<26:01, 21.50it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 91/33666 [00:04<25:59, 21.53it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 91/33666 [00:04<25:59, 21.52it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 92/33666 [00:04<25:56, 21.58it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 92/33666 [00:04<25:56, 21.57it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 93/33666 [00:04<25:52, 21.62it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 93/33666 [00:04<25:53, 21.62it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 94/33666 [00:04<25:48, 21.68it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 94/33666 [00:04<25:48, 21.68it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 95/33666 [00:04<25:46, 21.71it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 95/33666 [00:04<25:46, 21.71it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 96/33666 [00:04<25:42, 21.76it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 96/33666 [00:04<25:42, 21.76it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 97/33666 [00:04<25:40, 21.79it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 97/33666 [00:04<25:40, 21.79it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 98/33666 [00:04<25:37, 21.84it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 98/33666 [00:04<25:37, 21.84it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 99/33666 [00:04<25:33, 21.89it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 99/33666 [00:04<25:33, 21.89it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 100/33666 [00:04<25:30, 21.94it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 100/33666 [00:04<25:30, 21.94it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 101/33666 [00:04<25:26, 21.98it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 101/33666 [00:04<25:27, 21.98it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 102/33666 [00:04<25:23, 22.03it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 102/33666 [00:04<25:23, 22.03it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 103/33666 [00:04<25:20, 22.07it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 103/33666 [00:04<25:20, 22.07it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 104/33666 [00:04<25:16, 22.12it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 104/33666 [00:04<25:17, 22.12it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 105/33666 [00:04<25:16, 22.14it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 105/33666 [00:04<25:16, 22.13it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 106/33666 [00:04<25:13, 22.17it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 106/33666 [00:04<25:14, 22.17it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 107/33666 [00:04<25:11, 22.21it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 107/33666 [00:04<25:11, 22.21it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 108/33666 [00:04<25:07, 22.26it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 108/33666 [00:04<25:07, 22.26it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 109/33666 [00:04<25:05, 22.29it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 109/33666 [00:04<25:05, 22.29it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 110/33666 [00:04<25:02, 22.33it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 110/33666 [00:04<25:02, 22.33it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 111/33666 [00:04<24:59, 22.38it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 111/33666 [00:04<24:59, 22.38it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 112/33666 [00:04<24:57, 22.40it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 112/33666 [00:04<24:57, 22.40it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 113/33666 [00:05<24:57, 22.41it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 113/33666 [00:05<24:57, 22.41it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 114/33666 [00:05<24:54, 22.45it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 114/33666 [00:05<24:54, 22.45it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 115/33666 [00:05<24:51, 22.50it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 115/33666 [00:05<24:51, 22.50it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 116/33666 [00:05<24:49, 22.52it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 116/33666 [00:05<24:50, 22.52it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 117/33666 [00:05<24:46, 22.57it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 117/33666 [00:05<24:46, 22.56it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 118/33666 [00:05<24:43, 22.61it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 118/33666 [00:05<24:43, 22.61it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 119/33666 [00:05<24:41, 22.65it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 119/33666 [00:05<24:41, 22.64it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 120/33666 [00:05<24:38, 22.68it/s, loss=4.62, v_num=0]
Epoch 0:   0%|          | 120/33666 [00:05<24:39, 22.68it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 121/33666 [00:05<24:38, 22.70it/s, loss=4.61, v_num=0]
Epoch 0:   0%|          | 121/33666 [00:05<24:38, 22.69it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 122/33666 [00:05<24:37, 22.70it/s, loss=4.63, v_num=0]
Epoch 0:   0%|          | 122/33666 [00:05<24:37, 22.70it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 123/33666 [00:05<24:35, 22.74it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 123/33666 [00:05<24:35, 22.73it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 124/33666 [00:05<24:32, 22.77it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 124/33666 [00:05<24:33, 22.77it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 125/33666 [00:05<24:31, 22.80it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 125/33666 [00:05<24:31, 22.80it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 126/33666 [00:05<24:28, 22.83it/s, loss=4.68, v_num=0]
Epoch 0:   0%|          | 126/33666 [00:05<24:29, 22.83it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 127/33666 [00:05<24:26, 22.87it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 127/33666 [00:05<24:26, 22.87it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 128/33666 [00:05<24:24, 22.90it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 128/33666 [00:05<24:24, 22.90it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 129/33666 [00:05<24:24, 22.90it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 129/33666 [00:05<24:24, 22.89it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 130/33666 [00:05<24:22, 22.93it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 130/33666 [00:05<24:22, 22.92it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 131/33666 [00:05<24:20, 22.96it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 131/33666 [00:05<24:20, 22.96it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 132/33666 [00:05<24:18, 22.99it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 132/33666 [00:05<24:18, 22.99it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 133/33666 [00:05<24:15, 23.04it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 133/33666 [00:05<24:15, 23.04it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 134/33666 [00:05<24:15, 23.04it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 134/33666 [00:05<24:15, 23.04it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 135/33666 [00:05<24:13, 23.08it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 135/33666 [00:05<24:13, 23.07it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 136/33666 [00:05<24:11, 23.10it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 136/33666 [00:05<24:11, 23.10it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 137/33666 [00:05<24:09, 23.13it/s, loss=4.71, v_num=0]
Epoch 0:   0%|          | 137/33666 [00:05<24:09, 23.13it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 138/33666 [00:05<24:08, 23.15it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 138/33666 [00:05<24:08, 23.15it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 139/33666 [00:05<24:06, 23.18it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 139/33666 [00:05<24:06, 23.18it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 140/33666 [00:06<24:04, 23.20it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 140/33666 [00:06<24:05, 23.20it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 141/33666 [00:06<24:03, 23.23it/s, loss=4.7, v_num=0]
Epoch 0:   0%|          | 141/33666 [00:06<24:03, 23.23it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 142/33666 [00:06<24:01, 23.26it/s, loss=4.69, v_num=0]
Epoch 0:   0%|          | 142/33666 [00:06<24:01, 23.26it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 143/33666 [00:06<23:59, 23.29it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 143/33666 [00:06<23:59, 23.29it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 144/33666 [00:06<23:57, 23.33it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 144/33666 [00:06<23:57, 23.33it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 145/33666 [00:06<23:55, 23.35it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 145/33666 [00:06<23:55, 23.35it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 146/33666 [00:06<23:54, 23.36it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 146/33666 [00:06<23:54, 23.36it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 147/33666 [00:06<23:52, 23.39it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 147/33666 [00:06<23:52, 23.39it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 148/33666 [00:06<23:51, 23.42it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 148/33666 [00:06<23:51, 23.42it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 149/33666 [00:06<23:49, 23.45it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 149/33666 [00:06<23:49, 23.45it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 150/33666 [00:06<23:48, 23.47it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 150/33666 [00:06<23:48, 23.47it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 151/33666 [00:06<23:46, 23.50it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 151/33666 [00:06<23:46, 23.49it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 152/33666 [00:06<23:45, 23.51it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 152/33666 [00:06<23:45, 23.51it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 153/33666 [00:06<23:44, 23.53it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 153/33666 [00:06<23:44, 23.53it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 154/33666 [00:06<23:44, 23.53it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 154/33666 [00:06<23:44, 23.52it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 155/33666 [00:06<23:42, 23.55it/s, loss=4.67, v_num=0]
Epoch 0:   0%|          | 155/33666 [00:06<23:42, 23.55it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 156/33666 [00:06<23:41, 23.58it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 156/33666 [00:06<23:41, 23.58it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 157/33666 [00:06<23:39, 23.60it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 157/33666 [00:06<23:39, 23.60it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 158/33666 [00:06<23:38, 23.62it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 158/33666 [00:06<23:38, 23.62it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 159/33666 [00:06<23:38, 23.62it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 159/33666 [00:06<23:38, 23.62it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 160/33666 [00:06<23:36, 23.65it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 160/33666 [00:06<23:37, 23.65it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 161/33666 [00:06<23:35, 23.67it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 161/33666 [00:06<23:35, 23.66it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 162/33666 [00:06<23:33, 23.70it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 162/33666 [00:06<23:33, 23.70it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 163/33666 [00:06<23:33, 23.71it/s, loss=4.66, v_num=0]
Epoch 0:   0%|          | 163/33666 [00:06<23:33, 23.71it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 164/33666 [00:06<23:31, 23.73it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 164/33666 [00:06<23:31, 23.73it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 165/33666 [00:06<23:30, 23.75it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 165/33666 [00:06<23:30, 23.75it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 166/33666 [00:06<23:29, 23.76it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 166/33666 [00:06<23:29, 23.76it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 167/33666 [00:07<23:28, 23.78it/s, loss=4.64, v_num=0]
Epoch 0:   0%|          | 167/33666 [00:07<23:28, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 168/33666 [00:07<23:27, 23.81it/s, loss=4.65, v_num=0]
Epoch 0:   0%|          | 168/33666 [00:07<23:27, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 169/33666 [00:07<23:26, 23.82it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 169/33666 [00:07<23:26, 23.82it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 170/33666 [00:07<23:24, 23.85it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 170/33666 [00:07<23:24, 23.85it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 171/33666 [00:07<23:23, 23.86it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 171/33666 [00:07<23:24, 23.86it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 172/33666 [00:07<23:23, 23.86it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 172/33666 [00:07<23:23, 23.86it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 173/33666 [00:07<23:22, 23.88it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 173/33666 [00:07<23:22, 23.88it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 174/33666 [00:07<23:22, 23.88it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 174/33666 [00:07<23:22, 23.88it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 175/33666 [00:07<23:22, 23.88it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 175/33666 [00:07<23:22, 23.88it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 176/33666 [00:07<23:21, 23.90it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 176/33666 [00:07<23:21, 23.90it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 177/33666 [00:07<23:21, 23.90it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 177/33666 [00:07<23:21, 23.89it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 178/33666 [00:07<23:20, 23.91it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 178/33666 [00:07<23:20, 23.90it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 179/33666 [00:07<23:20, 23.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 179/33666 [00:07<23:20, 23.92it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 180/33666 [00:07<23:18, 23.94it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 180/33666 [00:07<23:18, 23.94it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 181/33666 [00:07<23:18, 23.94it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 181/33666 [00:07<23:18, 23.94it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 182/33666 [00:07<23:17, 23.96it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 182/33666 [00:07<23:17, 23.96it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 183/33666 [00:07<23:16, 23.97it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 183/33666 [00:07<23:16, 23.97it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 184/33666 [00:07<23:15, 23.99it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 184/33666 [00:07<23:15, 23.99it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 185/33666 [00:07<23:15, 23.99it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 185/33666 [00:07<23:15, 23.99it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 186/33666 [00:07<23:14, 24.01it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 186/33666 [00:07<23:14, 24.01it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 187/33666 [00:07<23:14, 24.02it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 187/33666 [00:07<23:14, 24.01it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 188/33666 [00:07<23:14, 24.01it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 188/33666 [00:07<23:14, 24.01it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 189/33666 [00:07<23:13, 24.02it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 189/33666 [00:07<23:13, 24.02it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 190/33666 [00:07<23:13, 24.02it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 190/33666 [00:07<23:13, 24.02it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 191/33666 [00:07<23:12, 24.04it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 191/33666 [00:07<23:12, 24.04it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 192/33666 [00:07<23:11, 24.06it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 192/33666 [00:07<23:11, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 193/33666 [00:08<23:12, 24.04it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 193/33666 [00:08<23:12, 24.03it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 194/33666 [00:08<23:11, 24.05it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 194/33666 [00:08<23:11, 24.05it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 195/33666 [00:08<23:11, 24.05it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 195/33666 [00:08<23:11, 24.05it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 196/33666 [00:08<23:10, 24.07it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 196/33666 [00:08<23:10, 24.07it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 197/33666 [00:08<23:10, 24.08it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 197/33666 [00:08<23:10, 24.08it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 198/33666 [00:08<23:09, 24.09it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 198/33666 [00:08<23:09, 24.09it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 199/33666 [00:08<23:08, 24.11it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 199/33666 [00:08<23:08, 24.10it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 200/33666 [00:08<23:07, 24.12it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 200/33666 [00:08<23:07, 24.12it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 201/33666 [00:08<23:06, 24.14it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 201/33666 [00:08<23:06, 24.14it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 202/33666 [00:08<23:06, 24.14it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 202/33666 [00:08<23:06, 24.13it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 203/33666 [00:08<23:06, 24.14it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 203/33666 [00:08<23:06, 24.14it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 204/33666 [00:08<23:06, 24.14it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 204/33666 [00:08<23:06, 24.14it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 205/33666 [00:08<23:05, 24.15it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 205/33666 [00:08<23:05, 24.15it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 206/33666 [00:08<23:04, 24.17it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 206/33666 [00:08<23:04, 24.17it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 207/33666 [00:08<23:02, 24.19it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 207/33666 [00:08<23:03, 24.19it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 208/33666 [00:08<23:01, 24.21it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 208/33666 [00:08<23:01, 24.21it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 209/33666 [00:08<23:01, 24.21it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 209/33666 [00:08<23:01, 24.21it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 210/33666 [00:08<23:00, 24.24it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 210/33666 [00:08<23:00, 24.23it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 211/33666 [00:08<22:59, 24.25it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 211/33666 [00:08<22:59, 24.25it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 212/33666 [00:08<22:58, 24.27it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 212/33666 [00:08<22:58, 24.27it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 213/33666 [00:08<22:57, 24.29it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 213/33666 [00:08<22:57, 24.29it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 214/33666 [00:08<22:55, 24.31it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 214/33666 [00:08<22:55, 24.31it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 215/33666 [00:08<22:54, 24.34it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 215/33666 [00:08<22:54, 24.34it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 216/33666 [00:08<22:53, 24.36it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 216/33666 [00:08<22:53, 24.36it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 217/33666 [00:08<22:51, 24.38it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 217/33666 [00:08<22:52, 24.38it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 218/33666 [00:08<22:50, 24.40it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 218/33666 [00:08<22:50, 24.40it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 219/33666 [00:08<22:49, 24.42it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 219/33666 [00:08<22:49, 24.42it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 220/33666 [00:09<22:49, 24.42it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 220/33666 [00:09<22:49, 24.42it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 221/33666 [00:09<22:49, 24.42it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 221/33666 [00:09<22:49, 24.42it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 222/33666 [00:09<22:49, 24.42it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 222/33666 [00:09<22:49, 24.42it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 223/33666 [00:09<22:48, 24.43it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 223/33666 [00:09<22:48, 24.43it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 224/33666 [00:09<22:48, 24.44it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 224/33666 [00:09<22:48, 24.44it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 225/33666 [00:09<22:48, 24.43it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 225/33666 [00:09<22:48, 24.43it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 226/33666 [00:09<22:48, 24.43it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 226/33666 [00:09<22:48, 24.43it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 227/33666 [00:09<22:49, 24.43it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 227/33666 [00:09<22:49, 24.42it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 228/33666 [00:09<22:49, 24.42it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 228/33666 [00:09<22:49, 24.42it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 229/33666 [00:09<22:48, 24.43it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 229/33666 [00:09<22:48, 24.43it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 230/33666 [00:09<22:48, 24.43it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 230/33666 [00:09<22:48, 24.43it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 231/33666 [00:09<22:49, 24.42it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 231/33666 [00:09<22:49, 24.42it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 232/33666 [00:09<22:49, 24.41it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 232/33666 [00:09<22:49, 24.41it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 233/33666 [00:09<22:49, 24.41it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 233/33666 [00:09<22:49, 24.40it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 234/33666 [00:09<22:50, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 234/33666 [00:09<22:50, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 235/33666 [00:09<22:50, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 235/33666 [00:09<22:50, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 236/33666 [00:09<22:50, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 236/33666 [00:09<22:50, 24.39it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 237/33666 [00:09<22:50, 24.40it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 237/33666 [00:09<22:50, 24.39it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 238/33666 [00:09<22:49, 24.41it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 238/33666 [00:09<22:49, 24.41it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 239/33666 [00:09<22:50, 24.40it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 239/33666 [00:09<22:50, 24.40it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 240/33666 [00:09<22:51, 24.38it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 240/33666 [00:09<22:51, 24.38it/s, loss=4.58, v_num=0]
Epoch 0:   1%|          | 241/33666 [00:09<22:52, 24.36it/s, loss=4.58, v_num=0]
Epoch 0:   1%|          | 241/33666 [00:09<22:52, 24.36it/s, loss=4.57, v_num=0]
Epoch 0:   1%|          | 242/33666 [00:09<22:53, 24.34it/s, loss=4.57, v_num=0]
Epoch 0:   1%|          | 242/33666 [00:09<22:53, 24.34it/s, loss=4.58, v_num=0]
Epoch 0:   1%|          | 243/33666 [00:09<22:53, 24.34it/s, loss=4.58, v_num=0]
Epoch 0:   1%|          | 243/33666 [00:09<22:53, 24.34it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 244/33666 [00:10<22:53, 24.33it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 244/33666 [00:10<22:53, 24.33it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 245/33666 [00:10<22:54, 24.31it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 245/33666 [00:10<22:54, 24.31it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 246/33666 [00:10<22:54, 24.32it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 246/33666 [00:10<22:54, 24.31it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 247/33666 [00:10<22:54, 24.31it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 247/33666 [00:10<22:54, 24.31it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 248/33666 [00:10<22:54, 24.31it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 248/33666 [00:10<22:54, 24.30it/s, loss=4.58, v_num=0]
Epoch 0:   1%|          | 249/33666 [00:10<22:55, 24.30it/s, loss=4.58, v_num=0]
Epoch 0:   1%|          | 249/33666 [00:10<22:55, 24.30it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 250/33666 [00:10<22:56, 24.28it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 250/33666 [00:10<22:56, 24.28it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 251/33666 [00:10<22:59, 24.22it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 251/33666 [00:10<22:59, 24.22it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 252/33666 [00:10<23:00, 24.21it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 252/33666 [00:10<23:00, 24.21it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 253/33666 [00:10<23:00, 24.20it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 253/33666 [00:10<23:00, 24.20it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 254/33666 [00:10<22:59, 24.22it/s, loss=4.59, v_num=0]
Epoch 0:   1%|          | 254/33666 [00:10<22:59, 24.22it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 255/33666 [00:10<22:59, 24.21it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 255/33666 [00:10<22:59, 24.21it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 256/33666 [00:10<23:00, 24.20it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 256/33666 [00:10<23:00, 24.20it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 257/33666 [00:10<23:01, 24.19it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 257/33666 [00:10<23:01, 24.19it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 258/33666 [00:10<23:02, 24.17it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 258/33666 [00:10<23:02, 24.17it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 259/33666 [00:10<23:02, 24.17it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 259/33666 [00:10<23:02, 24.17it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 260/33666 [00:10<23:03, 24.15it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 260/33666 [00:10<23:03, 24.15it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 261/33666 [00:10<23:03, 24.15it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 261/33666 [00:10<23:03, 24.15it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 262/33666 [00:10<23:03, 24.15it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 262/33666 [00:10<23:03, 24.15it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 263/33666 [00:10<23:03, 24.14it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 263/33666 [00:10<23:03, 24.14it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 264/33666 [00:10<23:04, 24.12it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 264/33666 [00:10<23:04, 24.12it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 265/33666 [00:10<23:04, 24.12it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 265/33666 [00:10<23:04, 24.12it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 266/33666 [00:11<23:05, 24.11it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 266/33666 [00:11<23:05, 24.11it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 267/33666 [00:11<23:06, 24.09it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 267/33666 [00:11<23:06, 24.09it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 268/33666 [00:11<23:06, 24.08it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 268/33666 [00:11<23:06, 24.08it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 269/33666 [00:11<23:08, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 269/33666 [00:11<23:08, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 270/33666 [00:11<23:08, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 270/33666 [00:11<23:08, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 271/33666 [00:11<23:08, 24.05it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 271/33666 [00:11<23:08, 24.05it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 272/33666 [00:11<23:09, 24.03it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 272/33666 [00:11<23:09, 24.03it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 273/33666 [00:11<23:09, 24.03it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 273/33666 [00:11<23:09, 24.03it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 274/33666 [00:11<23:10, 24.01it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 274/33666 [00:11<23:10, 24.01it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 275/33666 [00:11<23:11, 24.00it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 275/33666 [00:11<23:11, 24.00it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 276/33666 [00:11<23:12, 23.98it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 276/33666 [00:11<23:12, 23.98it/s, loss=4.67, v_num=0]
Epoch 0:   1%|          | 277/33666 [00:11<23:13, 23.97it/s, loss=4.67, v_num=0]
Epoch 0:   1%|          | 277/33666 [00:11<23:13, 23.97it/s, loss=4.67, v_num=0]
Epoch 0:   1%|          | 278/33666 [00:11<23:12, 23.97it/s, loss=4.67, v_num=0]
Epoch 0:   1%|          | 278/33666 [00:11<23:12, 23.97it/s, loss=4.68, v_num=0]
Epoch 0:   1%|          | 279/33666 [00:11<23:13, 23.96it/s, loss=4.68, v_num=0]
Epoch 0:   1%|          | 279/33666 [00:11<23:13, 23.96it/s, loss=4.68, v_num=0]
Epoch 0:   1%|          | 280/33666 [00:11<23:13, 23.95it/s, loss=4.68, v_num=0]
Epoch 0:   1%|          | 280/33666 [00:11<23:13, 23.95it/s, loss=4.67, v_num=0]
Epoch 0:   1%|          | 281/33666 [00:11<23:14, 23.94it/s, loss=4.67, v_num=0]
Epoch 0:   1%|          | 281/33666 [00:11<23:14, 23.94it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 282/33666 [00:11<23:15, 23.93it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 282/33666 [00:11<23:15, 23.93it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 283/33666 [00:11<23:15, 23.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 283/33666 [00:11<23:15, 23.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 284/33666 [00:11<23:15, 23.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 284/33666 [00:11<23:15, 23.91it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 285/33666 [00:11<23:16, 23.91it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 285/33666 [00:11<23:16, 23.90it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 286/33666 [00:11<23:16, 23.91it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 286/33666 [00:11<23:16, 23.91it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 287/33666 [00:12<23:16, 23.90it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 287/33666 [00:12<23:16, 23.90it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 288/33666 [00:12<23:17, 23.89it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 288/33666 [00:12<23:17, 23.89it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 289/33666 [00:12<23:17, 23.88it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 289/33666 [00:12<23:18, 23.87it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 290/33666 [00:12<23:18, 23.86it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 290/33666 [00:12<23:18, 23.86it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 291/33666 [00:12<23:19, 23.85it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 291/33666 [00:12<23:19, 23.85it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 292/33666 [00:12<23:19, 23.84it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 292/33666 [00:12<23:20, 23.84it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 293/33666 [00:12<23:20, 23.83it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 293/33666 [00:12<23:20, 23.82it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 294/33666 [00:12<23:21, 23.81it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 294/33666 [00:12<23:21, 23.81it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 295/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 295/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 296/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 296/33666 [00:12<23:22, 23.80it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 297/33666 [00:12<23:22, 23.80it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 297/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 298/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 298/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 299/33666 [00:12<23:21, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 299/33666 [00:12<23:22, 23.80it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 300/33666 [00:12<23:22, 23.80it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 300/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 301/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 301/33666 [00:12<23:22, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 302/33666 [00:12<23:22, 23.79it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 302/33666 [00:12<23:22, 23.79it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 303/33666 [00:12<23:22, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 303/33666 [00:12<23:23, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 304/33666 [00:12<23:22, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 304/33666 [00:12<23:22, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 305/33666 [00:12<23:22, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 305/33666 [00:12<23:22, 23.78it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 306/33666 [00:12<23:22, 23.78it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 306/33666 [00:12<23:22, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 307/33666 [00:12<23:22, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 307/33666 [00:12<23:22, 23.78it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 308/33666 [00:12<23:22, 23.79it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 308/33666 [00:12<23:22, 23.79it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 309/33666 [00:12<23:22, 23.79it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 309/33666 [00:12<23:22, 23.79it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 310/33666 [00:13<23:21, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 310/33666 [00:13<23:21, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 311/33666 [00:13<23:21, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 311/33666 [00:13<23:21, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 312/33666 [00:13<23:21, 23.80it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 312/33666 [00:13<23:21, 23.80it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 313/33666 [00:13<23:21, 23.80it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 313/33666 [00:13<23:21, 23.80it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 314/33666 [00:13<23:20, 23.82it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 314/33666 [00:13<23:20, 23.81it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 315/33666 [00:13<23:20, 23.82it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 315/33666 [00:13<23:20, 23.82it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 316/33666 [00:13<23:20, 23.82it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 316/33666 [00:13<23:20, 23.82it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 317/33666 [00:13<23:19, 23.82it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 317/33666 [00:13<23:19, 23.82it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 318/33666 [00:13<23:19, 23.83it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 318/33666 [00:13<23:19, 23.83it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 319/33666 [00:13<23:18, 23.84it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 319/33666 [00:13<23:18, 23.84it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 320/33666 [00:13<23:18, 23.84it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 320/33666 [00:13<23:18, 23.84it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 321/33666 [00:13<23:18, 23.84it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 321/33666 [00:13<23:18, 23.84it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 322/33666 [00:13<23:18, 23.85it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 322/33666 [00:13<23:18, 23.85it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 323/33666 [00:13<23:17, 23.87it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 323/33666 [00:13<23:17, 23.87it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 324/33666 [00:13<23:16, 23.88it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 324/33666 [00:13<23:16, 23.88it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 325/33666 [00:13<23:15, 23.89it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 325/33666 [00:13<23:15, 23.89it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 326/33666 [00:13<23:16, 23.88it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 326/33666 [00:13<23:16, 23.88it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 327/33666 [00:13<23:15, 23.89it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 327/33666 [00:13<23:15, 23.89it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 328/33666 [00:13<23:14, 23.90it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 328/33666 [00:13<23:14, 23.90it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 329/33666 [00:13<23:13, 23.92it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 329/33666 [00:13<23:13, 23.92it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 330/33666 [00:13<23:13, 23.92it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 330/33666 [00:13<23:13, 23.92it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 331/33666 [00:13<23:13, 23.93it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 331/33666 [00:13<23:13, 23.92it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 332/33666 [00:13<23:13, 23.93it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 332/33666 [00:13<23:13, 23.93it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 333/33666 [00:13<23:12, 23.93it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 333/33666 [00:13<23:12, 23.93it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 334/33666 [00:13<23:11, 23.95it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 334/33666 [00:13<23:11, 23.95it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 335/33666 [00:13<23:11, 23.95it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 335/33666 [00:13<23:12, 23.94it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 336/33666 [00:14<23:11, 23.96it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 336/33666 [00:14<23:11, 23.95it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 337/33666 [00:14<23:10, 23.97it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 337/33666 [00:14<23:10, 23.96it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 338/33666 [00:14<23:10, 23.97it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 338/33666 [00:14<23:10, 23.97it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 339/33666 [00:14<23:09, 23.99it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 339/33666 [00:14<23:09, 23.98it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 340/33666 [00:14<23:08, 24.00it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 340/33666 [00:14<23:08, 24.00it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 341/33666 [00:14<23:08, 24.00it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 341/33666 [00:14<23:08, 24.00it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 342/33666 [00:14<23:08, 24.00it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 342/33666 [00:14<23:08, 24.00it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 343/33666 [00:14<23:08, 24.00it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 343/33666 [00:14<23:08, 24.00it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 344/33666 [00:14<23:07, 24.01it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 344/33666 [00:14<23:07, 24.01it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 345/33666 [00:14<23:07, 24.02it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 345/33666 [00:14<23:07, 24.02it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 346/33666 [00:14<23:06, 24.03it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 346/33666 [00:14<23:06, 24.03it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 347/33666 [00:14<23:05, 24.04it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 347/33666 [00:14<23:06, 24.04it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 348/33666 [00:14<23:04, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 348/33666 [00:14<23:04, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 349/33666 [00:14<23:04, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 349/33666 [00:14<23:05, 24.06it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 350/33666 [00:14<23:04, 24.07it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 350/33666 [00:14<23:04, 24.07it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 351/33666 [00:14<23:03, 24.07it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 351/33666 [00:14<23:03, 24.07it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 352/33666 [00:14<23:03, 24.09it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 352/33666 [00:14<23:03, 24.09it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 353/33666 [00:14<23:02, 24.09it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 353/33666 [00:14<23:02, 24.09it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 354/33666 [00:14<23:02, 24.10it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 354/33666 [00:14<23:02, 24.10it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 355/33666 [00:14<23:01, 24.11it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 355/33666 [00:14<23:01, 24.10it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 356/33666 [00:14<23:01, 24.12it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 356/33666 [00:14<23:01, 24.12it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 357/33666 [00:14<23:00, 24.13it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 357/33666 [00:14<23:00, 24.13it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 358/33666 [00:14<22:59, 24.14it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 358/33666 [00:14<23:00, 24.14it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 359/33666 [00:14<22:59, 24.15it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 359/33666 [00:14<22:59, 24.15it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 360/33666 [00:14<22:58, 24.16it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 360/33666 [00:14<22:58, 24.16it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 361/33666 [00:14<22:58, 24.17it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 361/33666 [00:14<22:58, 24.17it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 362/33666 [00:14<22:57, 24.18it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 362/33666 [00:14<22:57, 24.18it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 363/33666 [00:15<22:56, 24.19it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 363/33666 [00:15<22:57, 24.18it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 364/33666 [00:15<22:56, 24.19it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 364/33666 [00:15<22:56, 24.19it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 365/33666 [00:15<22:56, 24.20it/s, loss=4.63, v_num=0]
Epoch 0:   1%|          | 365/33666 [00:15<22:56, 24.20it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 366/33666 [00:15<22:55, 24.21it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 366/33666 [00:15<22:55, 24.21it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 367/33666 [00:15<22:55, 24.22it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 367/33666 [00:15<22:55, 24.22it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 368/33666 [00:15<22:54, 24.23it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 368/33666 [00:15<22:54, 24.22it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 369/33666 [00:15<22:53, 24.23it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 369/33666 [00:15<22:53, 24.23it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 370/33666 [00:15<22:53, 24.24it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 370/33666 [00:15<22:53, 24.24it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 371/33666 [00:15<22:52, 24.25it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 371/33666 [00:15<22:52, 24.25it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 372/33666 [00:15<22:52, 24.26it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 372/33666 [00:15<22:52, 24.26it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 373/33666 [00:15<22:52, 24.26it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 373/33666 [00:15<22:52, 24.26it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 374/33666 [00:15<22:51, 24.27it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 374/33666 [00:15<22:51, 24.27it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 375/33666 [00:15<22:50, 24.29it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 375/33666 [00:15<22:50, 24.29it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 376/33666 [00:15<22:50, 24.29it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 376/33666 [00:15<22:50, 24.29it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 377/33666 [00:15<22:49, 24.30it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 377/33666 [00:15<22:49, 24.30it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 378/33666 [00:15<22:49, 24.30it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 378/33666 [00:15<22:49, 24.30it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 379/33666 [00:15<22:49, 24.31it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 379/33666 [00:15<22:49, 24.31it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 380/33666 [00:15<22:48, 24.31it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 380/33666 [00:15<22:49, 24.31it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 381/33666 [00:15<22:48, 24.32it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 381/33666 [00:15<22:48, 24.32it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 382/33666 [00:15<22:47, 24.33it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 382/33666 [00:15<22:47, 24.33it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 383/33666 [00:15<22:47, 24.33it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 383/33666 [00:15<22:47, 24.33it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 384/33666 [00:15<22:47, 24.34it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 384/33666 [00:15<22:47, 24.34it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 385/33666 [00:15<22:46, 24.36it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 385/33666 [00:15<22:46, 24.36it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 386/33666 [00:15<22:46, 24.36it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 386/33666 [00:15<22:46, 24.36it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 387/33666 [00:15<22:45, 24.37it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 387/33666 [00:15<22:45, 24.37it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 388/33666 [00:15<22:44, 24.38it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 388/33666 [00:15<22:44, 24.38it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 389/33666 [00:15<22:44, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 389/33666 [00:15<22:44, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 390/33666 [00:15<22:44, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 390/33666 [00:15<22:44, 24.39it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 391/33666 [00:16<22:43, 24.40it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 391/33666 [00:16<22:43, 24.40it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 392/33666 [00:16<22:43, 24.41it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 392/33666 [00:16<22:43, 24.41it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 393/33666 [00:16<22:42, 24.41it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 393/33666 [00:16<22:42, 24.41it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 394/33666 [00:16<22:42, 24.42it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 394/33666 [00:16<22:42, 24.42it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 395/33666 [00:16<22:41, 24.44it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 395/33666 [00:16<22:41, 24.44it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 396/33666 [00:16<22:40, 24.45it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 396/33666 [00:16<22:40, 24.45it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 397/33666 [00:16<22:40, 24.46it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 397/33666 [00:16<22:40, 24.46it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 398/33666 [00:16<22:39, 24.47it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 398/33666 [00:16<22:39, 24.47it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 399/33666 [00:16<22:39, 24.47it/s, loss=4.6, v_num=0]
Epoch 0:   1%|          | 399/33666 [00:16<22:39, 24.47it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 400/33666 [00:16<22:39, 24.48it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 400/33666 [00:16<22:39, 24.48it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 401/33666 [00:16<22:39, 24.47it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 401/33666 [00:16<22:39, 24.47it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 402/33666 [00:16<22:38, 24.48it/s, loss=4.61, v_num=0]
Epoch 0:   1%|          | 402/33666 [00:16<22:38, 24.48it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 403/33666 [00:16<22:38, 24.48it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 403/33666 [00:16<22:38, 24.48it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 404/33666 [00:16<22:38, 24.49it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 404/33666 [00:16<22:38, 24.49it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 405/33666 [00:16<22:37, 24.50it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 405/33666 [00:16<22:37, 24.49it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 406/33666 [00:16<22:37, 24.50it/s, loss=4.62, v_num=0]
Epoch 0:   1%|          | 406/33666 [00:16<22:37, 24.50it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 407/33666 [00:16<22:36, 24.51it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 407/33666 [00:16<22:37, 24.51it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 408/33666 [00:16<22:36, 24.52it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 408/33666 [00:16<22:36, 24.52it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 409/33666 [00:16<22:36, 24.52it/s, loss=4.64, v_num=0]
Epoch 0:   1%|          | 409/33666 [00:16<22:36, 24.52it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 410/33666 [00:16<22:35, 24.53it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 410/33666 [00:16<22:35, 24.53it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 411/33666 [00:16<22:34, 24.55it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 411/33666 [00:16<22:34, 24.55it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 412/33666 [00:16<22:34, 24.56it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 412/33666 [00:16<22:34, 24.56it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 413/33666 [00:16<22:33, 24.57it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 413/33666 [00:16<22:33, 24.57it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 414/33666 [00:16<22:32, 24.59it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 414/33666 [00:16<22:32, 24.58it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 415/33666 [00:16<22:31, 24.60it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 415/33666 [00:16<22:31, 24.60it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 416/33666 [00:16<22:30, 24.61it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 416/33666 [00:16<22:31, 24.61it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 417/33666 [00:16<22:30, 24.63it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 417/33666 [00:16<22:30, 24.62it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 418/33666 [00:16<22:29, 24.64it/s, loss=4.66, v_num=0]
Epoch 0:   1%|          | 418/33666 [00:16<22:29, 24.64it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 419/33666 [00:16<22:28, 24.65it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 419/33666 [00:16<22:28, 24.65it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 420/33666 [00:17<22:27, 24.67it/s, loss=4.65, v_num=0]
Epoch 0:   1%|          | 420/33666 [00:17<22:27, 24.66it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 421/33666 [00:17<22:27, 24.68it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 421/33666 [00:17<22:27, 24.68it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 422/33666 [00:17<22:26, 24.69it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 422/33666 [00:17<22:26, 24.69it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 423/33666 [00:17<22:25, 24.70it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 423/33666 [00:17<22:25, 24.70it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 424/33666 [00:17<22:25, 24.71it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 424/33666 [00:17<22:25, 24.71it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 425/33666 [00:17<22:24, 24.73it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 425/33666 [00:17<22:24, 24.73it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 426/33666 [00:17<22:23, 24.74it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 426/33666 [00:17<22:23, 24.74it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 427/33666 [00:17<22:22, 24.75it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 427/33666 [00:17<22:22, 24.75it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 428/33666 [00:17<22:22, 24.77it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 428/33666 [00:17<22:22, 24.77it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 429/33666 [00:17<22:21, 24.78it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 429/33666 [00:17<22:21, 24.78it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 430/33666 [00:17<22:20, 24.79it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 430/33666 [00:17<22:20, 24.79it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 431/33666 [00:17<22:20, 24.80it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 431/33666 [00:17<22:20, 24.80it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 432/33666 [00:17<22:19, 24.81it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 432/33666 [00:17<22:19, 24.81it/s, loss=4.6, v_num=0]
Epoch 0:   1%|▏         | 433/33666 [00:17<22:18, 24.83it/s, loss=4.6, v_num=0]
Epoch 0:   1%|▏         | 433/33666 [00:17<22:18, 24.83it/s, loss=4.6, v_num=0]
Epoch 0:   1%|▏         | 434/33666 [00:17<22:17, 24.84it/s, loss=4.6, v_num=0]
Epoch 0:   1%|▏         | 434/33666 [00:17<22:17, 24.84it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 435/33666 [00:17<22:17, 24.85it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 435/33666 [00:17<22:17, 24.85it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 436/33666 [00:17<22:16, 24.86it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 436/33666 [00:17<22:16, 24.86it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 437/33666 [00:17<22:15, 24.88it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 437/33666 [00:17<22:15, 24.87it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 438/33666 [00:17<22:15, 24.89it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 438/33666 [00:17<22:15, 24.89it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 439/33666 [00:17<22:14, 24.90it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 439/33666 [00:17<22:14, 24.90it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 440/33666 [00:17<22:13, 24.91it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 440/33666 [00:17<22:13, 24.91it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 441/33666 [00:17<22:13, 24.92it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 441/33666 [00:17<22:13, 24.92it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 442/33666 [00:17<22:13, 24.92it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 442/33666 [00:17<22:13, 24.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 443/33666 [00:17<22:13, 24.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 443/33666 [00:17<22:13, 24.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 444/33666 [00:17<22:13, 24.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 444/33666 [00:17<22:13, 24.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 445/33666 [00:17<22:12, 24.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 445/33666 [00:17<22:12, 24.92it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 446/33666 [00:17<22:12, 24.93it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 446/33666 [00:17<22:12, 24.93it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 447/33666 [00:17<22:12, 24.93it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 447/33666 [00:17<22:12, 24.93it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 448/33666 [00:17<22:11, 24.94it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 448/33666 [00:17<22:11, 24.94it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 449/33666 [00:17<22:11, 24.95it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 449/33666 [00:17<22:11, 24.95it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 450/33666 [00:18<22:10, 24.96it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 450/33666 [00:18<22:10, 24.96it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 451/33666 [00:18<22:10, 24.97it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 451/33666 [00:18<22:10, 24.97it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 452/33666 [00:18<22:09, 24.98it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 452/33666 [00:18<22:09, 24.98it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 453/33666 [00:18<22:08, 24.99it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 453/33666 [00:18<22:08, 24.99it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 454/33666 [00:18<22:08, 25.00it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 454/33666 [00:18<22:08, 25.00it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 455/33666 [00:18<22:07, 25.02it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 455/33666 [00:18<22:07, 25.02it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 456/33666 [00:18<22:06, 25.03it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 456/33666 [00:18<22:06, 25.03it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 457/33666 [00:18<22:06, 25.04it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 457/33666 [00:18<22:06, 25.04it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 458/33666 [00:18<22:06, 25.04it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 458/33666 [00:18<22:06, 25.04it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 459/33666 [00:18<22:05, 25.05it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 459/33666 [00:18<22:05, 25.05it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 460/33666 [00:18<22:05, 25.05it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 460/33666 [00:18<22:05, 25.05it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 461/33666 [00:18<22:05, 25.06it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 461/33666 [00:18<22:05, 25.06it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 462/33666 [00:18<22:04, 25.06it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 462/33666 [00:18<22:05, 25.06it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 463/33666 [00:18<22:04, 25.07it/s, loss=4.61, v_num=0]
Epoch 0:   1%|▏         | 463/33666 [00:18<22:04, 25.07it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 464/33666 [00:18<22:04, 25.07it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 464/33666 [00:18<22:04, 25.07it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 465/33666 [00:18<22:03, 25.08it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 465/33666 [00:18<22:03, 25.08it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 466/33666 [00:18<22:03, 25.09it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 466/33666 [00:18<22:03, 25.09it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 467/33666 [00:18<22:02, 25.10it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 467/33666 [00:18<22:02, 25.10it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 468/33666 [00:18<22:02, 25.10it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 468/33666 [00:18<22:02, 25.10it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 469/33666 [00:18<22:01, 25.11it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 469/33666 [00:18<22:01, 25.11it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 470/33666 [00:18<22:01, 25.12it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 470/33666 [00:18<22:01, 25.12it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 471/33666 [00:18<22:00, 25.13it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 471/33666 [00:18<22:00, 25.13it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 472/33666 [00:18<22:00, 25.13it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 472/33666 [00:18<22:00, 25.13it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 473/33666 [00:18<22:00, 25.14it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 473/33666 [00:18<22:00, 25.14it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 474/33666 [00:18<21:59, 25.15it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 474/33666 [00:18<21:59, 25.15it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 475/33666 [00:18<21:59, 25.16it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 475/33666 [00:18<21:59, 25.16it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 476/33666 [00:18<21:58, 25.17it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 476/33666 [00:18<21:58, 25.17it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 477/33666 [00:18<21:58, 25.18it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 477/33666 [00:18<21:58, 25.18it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 478/33666 [00:18<21:57, 25.19it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 478/33666 [00:18<21:57, 25.19it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 479/33666 [00:19<21:56, 25.20it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 479/33666 [00:19<21:56, 25.20it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 480/33666 [00:19<21:56, 25.21it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 480/33666 [00:19<21:56, 25.21it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 481/33666 [00:19<21:55, 25.22it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 481/33666 [00:19<21:55, 25.22it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 482/33666 [00:19<21:55, 25.23it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 482/33666 [00:19<21:55, 25.23it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 483/33666 [00:19<21:54, 25.24it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 483/33666 [00:19<21:54, 25.24it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 484/33666 [00:19<21:53, 25.25it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 484/33666 [00:19<21:53, 25.25it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 485/33666 [00:19<21:53, 25.26it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 485/33666 [00:19<21:53, 25.26it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 486/33666 [00:19<21:52, 25.27it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 486/33666 [00:19<21:52, 25.27it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 487/33666 [00:19<21:52, 25.28it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 487/33666 [00:19<21:52, 25.28it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 488/33666 [00:19<21:51, 25.29it/s, loss=4.62, v_num=0]
Epoch 0:   1%|▏         | 488/33666 [00:19<21:51, 25.29it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 489/33666 [00:19<21:51, 25.30it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 489/33666 [00:19<21:51, 25.30it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 490/33666 [00:19<21:51, 25.30it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 490/33666 [00:19<21:51, 25.30it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 491/33666 [00:19<21:51, 25.30it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 491/33666 [00:19<21:51, 25.30it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 492/33666 [00:19<21:50, 25.31it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 492/33666 [00:19<21:50, 25.31it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 493/33666 [00:19<21:50, 25.32it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 493/33666 [00:19<21:50, 25.31it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 494/33666 [00:19<21:50, 25.32it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 494/33666 [00:19<21:50, 25.32it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 495/33666 [00:19<21:49, 25.32it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 495/33666 [00:19<21:49, 25.32it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 496/33666 [00:19<21:49, 25.33it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 496/33666 [00:19<21:49, 25.33it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 497/33666 [00:19<21:49, 25.33it/s, loss=4.65, v_num=0]
Epoch 0:   1%|▏         | 497/33666 [00:19<21:49, 25.33it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 498/33666 [00:19<21:48, 25.34it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 498/33666 [00:19<21:48, 25.34it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 499/33666 [00:19<21:48, 25.35it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 499/33666 [00:19<21:48, 25.35it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 500/33666 [00:19<21:48, 25.35it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 500/33666 [00:19<21:48, 25.35it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 501/33666 [00:19<21:48, 25.35it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 501/33666 [00:19<21:48, 25.35it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 502/33666 [00:19<21:48, 25.35it/s, loss=4.64, v_num=0]
Epoch 0:   1%|▏         | 502/33666 [00:19<21:48, 25.35it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 503/33666 [00:19<21:48, 25.35it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 503/33666 [00:19<21:48, 25.35it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 504/33666 [00:19<21:47, 25.36it/s, loss=4.63, v_num=0]
Epoch 0:   1%|▏         | 504/33666 [00:19<21:47, 25.36it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 505/33666 [00:19<21:47, 25.36it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 505/33666 [00:19<21:47, 25.36it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 506/33666 [00:19<21:47, 25.37it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 506/33666 [00:19<21:47, 25.37it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 507/33666 [00:19<21:46, 25.37it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 507/33666 [00:19<21:46, 25.37it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 508/33666 [00:20<21:46, 25.38it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 508/33666 [00:20<21:46, 25.38it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 509/33666 [00:20<21:46, 25.38it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 509/33666 [00:20<21:46, 25.38it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 510/33666 [00:20<21:45, 25.39it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 510/33666 [00:20<21:45, 25.39it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 511/33666 [00:20<21:45, 25.40it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 511/33666 [00:20<21:45, 25.40it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 512/33666 [00:20<21:44, 25.41it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 512/33666 [00:20<21:44, 25.41it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 513/33666 [00:20<21:44, 25.42it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 513/33666 [00:20<21:44, 25.42it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 514/33666 [00:20<21:44, 25.42it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 514/33666 [00:20<21:44, 25.42it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 515/33666 [00:20<21:43, 25.43it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 515/33666 [00:20<21:43, 25.43it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 516/33666 [00:20<21:43, 25.44it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 516/33666 [00:20<21:43, 25.44it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 517/33666 [00:20<21:42, 25.45it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 517/33666 [00:20<21:42, 25.45it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 518/33666 [00:20<21:42, 25.46it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 518/33666 [00:20<21:42, 25.46it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 519/33666 [00:20<21:41, 25.46it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 519/33666 [00:20<21:41, 25.46it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 520/33666 [00:20<21:41, 25.47it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 520/33666 [00:20<21:41, 25.47it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 521/33666 [00:20<21:40, 25.48it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 521/33666 [00:20<21:40, 25.48it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 522/33666 [00:20<21:40, 25.49it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 522/33666 [00:20<21:40, 25.49it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 523/33666 [00:20<21:39, 25.50it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 523/33666 [00:20<21:39, 25.50it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 524/33666 [00:20<21:39, 25.51it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 524/33666 [00:20<21:39, 25.51it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 525/33666 [00:20<21:39, 25.51it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 525/33666 [00:20<21:39, 25.51it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 526/33666 [00:20<21:38, 25.52it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 526/33666 [00:20<21:38, 25.52it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 527/33666 [00:20<21:38, 25.52it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 527/33666 [00:20<21:38, 25.52it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 528/33666 [00:20<21:37, 25.53it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 528/33666 [00:20<21:38, 25.53it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 529/33666 [00:20<21:37, 25.54it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 529/33666 [00:20<21:37, 25.54it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 530/33666 [00:20<21:36, 25.55it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 530/33666 [00:20<21:36, 25.55it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 531/33666 [00:20<21:36, 25.56it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 531/33666 [00:20<21:36, 25.56it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 532/33666 [00:20<21:35, 25.57it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 532/33666 [00:20<21:35, 25.57it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 533/33666 [00:20<21:35, 25.58it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 533/33666 [00:20<21:35, 25.58it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 534/33666 [00:20<21:34, 25.59it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 534/33666 [00:20<21:34, 25.59it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 535/33666 [00:20<21:34, 25.60it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 535/33666 [00:20<21:34, 25.60it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 536/33666 [00:20<21:33, 25.60it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 536/33666 [00:20<21:34, 25.60it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 537/33666 [00:20<21:33, 25.61it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 537/33666 [00:20<21:33, 25.61it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 538/33666 [00:20<21:32, 25.62it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 538/33666 [00:20<21:32, 25.62it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 539/33666 [00:21<21:32, 25.63it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 539/33666 [00:21<21:32, 25.63it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 540/33666 [00:21<21:31, 25.64it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 540/33666 [00:21<21:31, 25.64it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 541/33666 [00:21<21:31, 25.65it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 541/33666 [00:21<21:31, 25.65it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 542/33666 [00:21<21:31, 25.66it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 542/33666 [00:21<21:31, 25.66it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 543/33666 [00:21<21:30, 25.67it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 543/33666 [00:21<21:30, 25.67it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 544/33666 [00:21<21:30, 25.67it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 544/33666 [00:21<21:30, 25.67it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 545/33666 [00:21<21:29, 25.68it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 545/33666 [00:21<21:29, 25.68it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 546/33666 [00:21<21:29, 25.69it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 546/33666 [00:21<21:29, 25.69it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 547/33666 [00:21<21:28, 25.70it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 547/33666 [00:21<21:28, 25.70it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 548/33666 [00:21<21:28, 25.71it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 548/33666 [00:21<21:28, 25.71it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 549/33666 [00:21<21:27, 25.72it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 549/33666 [00:21<21:27, 25.72it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 550/33666 [00:21<21:27, 25.73it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 550/33666 [00:21<21:27, 25.73it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 551/33666 [00:21<21:26, 25.73it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 551/33666 [00:21<21:26, 25.73it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 552/33666 [00:21<21:26, 25.74it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 552/33666 [00:21<21:26, 25.74it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 553/33666 [00:21<21:25, 25.75it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 553/33666 [00:21<21:25, 25.75it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 554/33666 [00:21<21:25, 25.76it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 554/33666 [00:21<21:25, 25.76it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 555/33666 [00:21<21:24, 25.77it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 555/33666 [00:21<21:24, 25.77it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 556/33666 [00:21<21:24, 25.78it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 556/33666 [00:21<21:24, 25.78it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 557/33666 [00:21<21:23, 25.79it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 557/33666 [00:21<21:23, 25.79it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 558/33666 [00:21<21:23, 25.79it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 558/33666 [00:21<21:23, 25.79it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 559/33666 [00:21<21:23, 25.80it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 559/33666 [00:21<21:23, 25.80it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 560/33666 [00:21<21:22, 25.81it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 560/33666 [00:21<21:22, 25.81it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 561/33666 [00:21<21:22, 25.82it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 561/33666 [00:21<21:22, 25.82it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 562/33666 [00:21<21:21, 25.83it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 562/33666 [00:21<21:21, 25.83it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 563/33666 [00:21<21:21, 25.84it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 563/33666 [00:21<21:21, 25.84it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 564/33666 [00:21<21:20, 25.85it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 564/33666 [00:21<21:20, 25.84it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 565/33666 [00:21<21:20, 25.85it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 565/33666 [00:21<21:20, 25.85it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 566/33666 [00:21<21:19, 25.86it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 566/33666 [00:21<21:19, 25.86it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 567/33666 [00:21<21:19, 25.87it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 567/33666 [00:21<21:19, 25.87it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 568/33666 [00:21<21:19, 25.88it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 568/33666 [00:21<21:19, 25.88it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 569/33666 [00:21<21:18, 25.88it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 569/33666 [00:21<21:18, 25.88it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 570/33666 [00:22<21:18, 25.89it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 570/33666 [00:22<21:18, 25.89it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 571/33666 [00:22<21:17, 25.90it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 571/33666 [00:22<21:17, 25.90it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 572/33666 [00:22<21:17, 25.91it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 572/33666 [00:22<21:17, 25.91it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 573/33666 [00:22<21:16, 25.92it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 573/33666 [00:22<21:17, 25.91it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 574/33666 [00:22<21:16, 25.92it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 574/33666 [00:22<21:16, 25.92it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 575/33666 [00:22<21:16, 25.93it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 575/33666 [00:22<21:16, 25.93it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 576/33666 [00:22<21:15, 25.94it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 576/33666 [00:22<21:15, 25.94it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 577/33666 [00:22<21:15, 25.95it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 577/33666 [00:22<21:15, 25.95it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 578/33666 [00:22<21:14, 25.95it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 578/33666 [00:22<21:14, 25.95it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 579/33666 [00:22<21:14, 25.96it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 579/33666 [00:22<21:14, 25.96it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 580/33666 [00:22<21:13, 25.97it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 580/33666 [00:22<21:13, 25.97it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 581/33666 [00:22<21:13, 25.97it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 581/33666 [00:22<21:13, 25.97it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 582/33666 [00:22<21:13, 25.97it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 582/33666 [00:22<21:13, 25.97it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 583/33666 [00:22<21:13, 25.97it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 583/33666 [00:22<21:13, 25.97it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 584/33666 [00:22<21:13, 25.97it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 584/33666 [00:22<21:13, 25.97it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 585/33666 [00:22<21:13, 25.97it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 585/33666 [00:22<21:13, 25.97it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 586/33666 [00:22<21:13, 25.98it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 586/33666 [00:22<21:13, 25.98it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 587/33666 [00:22<21:13, 25.98it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 587/33666 [00:22<21:13, 25.98it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 588/33666 [00:22<21:13, 25.98it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 588/33666 [00:22<21:13, 25.98it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 589/33666 [00:22<21:12, 25.98it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 589/33666 [00:22<21:12, 25.98it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 590/33666 [00:22<21:12, 25.99it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 590/33666 [00:22<21:12, 25.99it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 591/33666 [00:22<21:12, 25.99it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 591/33666 [00:22<21:12, 25.99it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 592/33666 [00:22<21:12, 25.99it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 592/33666 [00:22<21:12, 25.98it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 593/33666 [00:22<21:12, 25.99it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 593/33666 [00:22<21:12, 25.99it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 594/33666 [00:22<21:12, 25.99it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 594/33666 [00:22<21:12, 25.99it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 595/33666 [00:22<21:12, 25.99it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 595/33666 [00:22<21:12, 25.99it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 596/33666 [00:22<21:12, 26.00it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 596/33666 [00:22<21:12, 26.00it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 597/33666 [00:22<21:11, 26.00it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 597/33666 [00:22<21:11, 26.00it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 598/33666 [00:22<21:11, 26.00it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 598/33666 [00:22<21:11, 26.00it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 599/33666 [00:23<21:11, 26.00it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 599/33666 [00:23<21:11, 26.00it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 600/33666 [00:23<21:11, 26.01it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 600/33666 [00:23<21:11, 26.01it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 601/33666 [00:23<21:11, 26.00it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 601/33666 [00:23<21:11, 26.00it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 602/33666 [00:23<21:11, 26.00it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 602/33666 [00:23<21:11, 26.00it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 603/33666 [00:23<21:11, 26.01it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 603/33666 [00:23<21:11, 26.01it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 604/33666 [00:23<21:11, 26.01it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 604/33666 [00:23<21:11, 26.01it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 605/33666 [00:23<21:10, 26.01it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 605/33666 [00:23<21:10, 26.01it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 606/33666 [00:23<21:10, 26.01it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 606/33666 [00:23<21:10, 26.01it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 607/33666 [00:23<21:10, 26.02it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 607/33666 [00:23<21:10, 26.02it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 608/33666 [00:23<21:10, 26.02it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 608/33666 [00:23<21:10, 26.02it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 609/33666 [00:23<21:10, 26.02it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 609/33666 [00:23<21:10, 26.02it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 610/33666 [00:23<21:10, 26.03it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 610/33666 [00:23<21:10, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 611/33666 [00:23<21:10, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 611/33666 [00:23<21:10, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 612/33666 [00:23<21:09, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 612/33666 [00:23<21:09, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 613/33666 [00:23<21:09, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 613/33666 [00:23<21:09, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 614/33666 [00:23<21:09, 26.04it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 614/33666 [00:23<21:09, 26.04it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 615/33666 [00:23<21:09, 26.04it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 615/33666 [00:23<21:09, 26.04it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 616/33666 [00:23<21:09, 26.04it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 616/33666 [00:23<21:09, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 617/33666 [00:23<21:09, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 617/33666 [00:23<21:09, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 618/33666 [00:23<21:08, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 618/33666 [00:23<21:08, 26.04it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 619/33666 [00:23<21:08, 26.05it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 619/33666 [00:23<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 620/33666 [00:23<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 620/33666 [00:23<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 621/33666 [00:23<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 621/33666 [00:23<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 622/33666 [00:23<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 622/33666 [00:23<21:08, 26.05it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 623/33666 [00:23<21:08, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 623/33666 [00:23<21:08, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 624/33666 [00:23<21:07, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 624/33666 [00:23<21:07, 26.06it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 625/33666 [00:23<21:07, 26.06it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 625/33666 [00:23<21:07, 26.06it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 626/33666 [00:24<21:07, 26.06it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 626/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 627/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 627/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 628/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 628/33666 [00:24<21:07, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 629/33666 [00:24<21:07, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 629/33666 [00:24<21:07, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 630/33666 [00:24<21:07, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 630/33666 [00:24<21:07, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 631/33666 [00:24<21:07, 26.06it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 631/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 632/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 632/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 633/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 633/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 634/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 634/33666 [00:24<21:07, 26.06it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 635/33666 [00:24<21:07, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 635/33666 [00:24<21:07, 26.05it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 636/33666 [00:24<21:07, 26.05it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 636/33666 [00:24<21:07, 26.05it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 637/33666 [00:24<21:07, 26.05it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 637/33666 [00:24<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 638/33666 [00:24<21:07, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 638/33666 [00:24<21:07, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 639/33666 [00:24<21:08, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 639/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 640/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 640/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 641/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 641/33666 [00:24<21:08, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 642/33666 [00:24<21:08, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 642/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 643/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 643/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 644/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 644/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 645/33666 [00:24<21:08, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 645/33666 [00:24<21:08, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 646/33666 [00:24<21:08, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 646/33666 [00:24<21:08, 26.04it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 647/33666 [00:24<21:08, 26.04it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 647/33666 [00:24<21:08, 26.04it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 648/33666 [00:24<21:08, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 648/33666 [00:24<21:08, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 649/33666 [00:24<21:08, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 649/33666 [00:24<21:08, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 650/33666 [00:24<21:08, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 650/33666 [00:24<21:08, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 651/33666 [00:25<21:08, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 651/33666 [00:25<21:08, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 652/33666 [00:25<21:08, 26.03it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 652/33666 [00:25<21:08, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 653/33666 [00:25<21:08, 26.02it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 653/33666 [00:25<21:08, 26.02it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 654/33666 [00:25<21:08, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 654/33666 [00:25<21:08, 26.03it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 655/33666 [00:25<21:08, 26.02it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 655/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 656/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 656/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 657/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 657/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 658/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 658/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 659/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 659/33666 [00:25<21:08, 26.02it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 660/33666 [00:25<21:08, 26.02it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 660/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 661/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 661/33666 [00:25<21:08, 26.02it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 662/33666 [00:25<21:08, 26.01it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 662/33666 [00:25<21:08, 26.01it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 663/33666 [00:25<21:08, 26.01it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 663/33666 [00:25<21:08, 26.01it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 664/33666 [00:25<21:08, 26.02it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 664/33666 [00:25<21:08, 26.02it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 665/33666 [00:25<21:08, 26.01it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 665/33666 [00:25<21:08, 26.01it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 666/33666 [00:25<21:08, 26.02it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 666/33666 [00:25<21:08, 26.02it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 667/33666 [00:25<21:08, 26.02it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 667/33666 [00:25<21:08, 26.02it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 668/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 668/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 669/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 669/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 670/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 670/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 671/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 671/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 672/33666 [00:25<21:07, 26.03it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 672/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 673/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 673/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 674/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 674/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 675/33666 [00:25<21:07, 26.04it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 675/33666 [00:25<21:07, 26.04it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 676/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 676/33666 [00:25<21:07, 26.03it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 677/33666 [00:26<21:07, 26.04it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 677/33666 [00:26<21:07, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 678/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 678/33666 [00:26<21:06, 26.04it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 679/33666 [00:26<21:06, 26.04it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 679/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 680/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 680/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 681/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 681/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 682/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 682/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 683/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 683/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 684/33666 [00:26<21:06, 26.05it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 684/33666 [00:26<21:06, 26.05it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 685/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 685/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 686/33666 [00:26<21:06, 26.05it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 686/33666 [00:26<21:06, 26.05it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 687/33666 [00:26<21:06, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 687/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 688/33666 [00:26<21:06, 26.04it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 688/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 689/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 689/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 690/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 690/33666 [00:26<21:06, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 691/33666 [00:26<21:06, 26.04it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 691/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 692/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 692/33666 [00:26<21:06, 26.04it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 693/33666 [00:26<21:05, 26.05it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 693/33666 [00:26<21:05, 26.05it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 694/33666 [00:26<21:05, 26.05it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 694/33666 [00:26<21:05, 26.05it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 695/33666 [00:26<21:05, 26.05it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 695/33666 [00:26<21:05, 26.05it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 696/33666 [00:26<21:05, 26.05it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 696/33666 [00:26<21:05, 26.05it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 697/33666 [00:26<21:05, 26.06it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 697/33666 [00:26<21:05, 26.06it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 698/33666 [00:26<21:05, 26.06it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 698/33666 [00:26<21:05, 26.06it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 699/33666 [00:26<21:04, 26.06it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 699/33666 [00:26<21:04, 26.06it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 700/33666 [00:26<21:04, 26.06it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 700/33666 [00:26<21:04, 26.06it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 701/33666 [00:26<21:04, 26.06it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 701/33666 [00:26<21:04, 26.06it/s, loss=4.69, v_num=0]
Epoch 0:   2%|▏         | 702/33666 [00:26<21:04, 26.07it/s, loss=4.69, v_num=0]
Epoch 0:   2%|▏         | 702/33666 [00:26<21:04, 26.06it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 703/33666 [00:26<21:04, 26.07it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 703/33666 [00:26<21:04, 26.07it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 704/33666 [00:27<21:04, 26.07it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 704/33666 [00:27<21:04, 26.07it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 705/33666 [00:27<21:04, 26.07it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 705/33666 [00:27<21:04, 26.07it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 706/33666 [00:27<21:03, 26.08it/s, loss=4.7, v_num=0]
Epoch 0:   2%|▏         | 706/33666 [00:27<21:03, 26.08it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 707/33666 [00:27<21:04, 26.07it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 707/33666 [00:27<21:04, 26.07it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 708/33666 [00:27<21:03, 26.08it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 708/33666 [00:27<21:03, 26.08it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 709/33666 [00:27<21:03, 26.08it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 709/33666 [00:27<21:03, 26.08it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 710/33666 [00:27<21:03, 26.07it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 710/33666 [00:27<21:03, 26.07it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 711/33666 [00:27<21:03, 26.07it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 711/33666 [00:27<21:03, 26.07it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 712/33666 [00:27<21:03, 26.07it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 712/33666 [00:27<21:03, 26.07it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 713/33666 [00:27<21:03, 26.08it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 713/33666 [00:27<21:03, 26.08it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 714/33666 [00:27<21:03, 26.08it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 714/33666 [00:27<21:03, 26.08it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 715/33666 [00:27<21:03, 26.08it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 715/33666 [00:27<21:03, 26.08it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 716/33666 [00:27<21:03, 26.09it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 716/33666 [00:27<21:03, 26.09it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 717/33666 [00:27<21:02, 26.09it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 717/33666 [00:27<21:02, 26.09it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 718/33666 [00:27<21:02, 26.09it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 718/33666 [00:27<21:02, 26.09it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 719/33666 [00:27<21:02, 26.09it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 719/33666 [00:27<21:02, 26.09it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 720/33666 [00:27<21:02, 26.10it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 720/33666 [00:27<21:02, 26.10it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 721/33666 [00:27<21:02, 26.10it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 721/33666 [00:27<21:02, 26.10it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 722/33666 [00:27<21:01, 26.11it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 722/33666 [00:27<21:01, 26.11it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 723/33666 [00:27<21:01, 26.10it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 723/33666 [00:27<21:01, 26.10it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 724/33666 [00:27<21:02, 26.10it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 724/33666 [00:27<21:02, 26.10it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 725/33666 [00:27<21:01, 26.10it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 725/33666 [00:27<21:01, 26.10it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 726/33666 [00:27<21:02, 26.10it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 726/33666 [00:27<21:02, 26.10it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 727/33666 [00:27<21:02, 26.10it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 727/33666 [00:27<21:02, 26.10it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 728/33666 [00:27<21:02, 26.10it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 728/33666 [00:27<21:02, 26.10it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 729/33666 [00:27<21:02, 26.10it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 729/33666 [00:27<21:02, 26.10it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 730/33666 [00:27<21:01, 26.10it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 730/33666 [00:27<21:01, 26.10it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 731/33666 [00:28<21:01, 26.10it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 731/33666 [00:28<21:01, 26.10it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 732/33666 [00:28<21:01, 26.10it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 732/33666 [00:28<21:01, 26.10it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 733/33666 [00:28<21:01, 26.11it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 733/33666 [00:28<21:01, 26.11it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 734/33666 [00:28<21:01, 26.10it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 734/33666 [00:28<21:01, 26.10it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 735/33666 [00:28<21:01, 26.11it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 735/33666 [00:28<21:01, 26.10it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 736/33666 [00:28<21:01, 26.10it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 736/33666 [00:28<21:01, 26.10it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 737/33666 [00:28<21:01, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 737/33666 [00:28<21:01, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 738/33666 [00:28<21:00, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 738/33666 [00:28<21:00, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 739/33666 [00:28<21:00, 26.12it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 739/33666 [00:28<21:00, 26.12it/s, loss=4.69, v_num=0]
Epoch 0:   2%|▏         | 740/33666 [00:28<21:00, 26.12it/s, loss=4.69, v_num=0]
Epoch 0:   2%|▏         | 740/33666 [00:28<21:00, 26.12it/s, loss=4.69, v_num=0]
Epoch 0:   2%|▏         | 741/33666 [00:28<21:00, 26.12it/s, loss=4.69, v_num=0]
Epoch 0:   2%|▏         | 741/33666 [00:28<21:00, 26.12it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 742/33666 [00:28<21:00, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 742/33666 [00:28<21:00, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 743/33666 [00:28<21:00, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 743/33666 [00:28<21:00, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 744/33666 [00:28<21:00, 26.12it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 744/33666 [00:28<21:00, 26.11it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 745/33666 [00:28<21:00, 26.12it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 745/33666 [00:28<21:00, 26.11it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 746/33666 [00:28<21:00, 26.11it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 746/33666 [00:28<21:00, 26.11it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 747/33666 [00:28<21:00, 26.11it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 747/33666 [00:28<21:00, 26.11it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 748/33666 [00:28<21:00, 26.12it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 748/33666 [00:28<21:00, 26.11it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 749/33666 [00:28<21:00, 26.12it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 749/33666 [00:28<21:00, 26.12it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 750/33666 [00:28<21:00, 26.12it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 750/33666 [00:28<21:00, 26.12it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 751/33666 [00:28<21:00, 26.11it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 751/33666 [00:28<21:00, 26.11it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 752/33666 [00:28<21:00, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 752/33666 [00:28<21:00, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 753/33666 [00:28<20:59, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 753/33666 [00:28<20:59, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 754/33666 [00:28<21:00, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 754/33666 [00:28<21:00, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 755/33666 [00:28<21:00, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 755/33666 [00:28<21:00, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 756/33666 [00:28<20:59, 26.12it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 756/33666 [00:28<20:59, 26.12it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 757/33666 [00:28<20:59, 26.12it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 757/33666 [00:28<20:59, 26.12it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 758/33666 [00:29<20:59, 26.12it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 758/33666 [00:29<20:59, 26.12it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 759/33666 [00:29<20:59, 26.12it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 759/33666 [00:29<20:59, 26.12it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 760/33666 [00:29<20:59, 26.13it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 760/33666 [00:29<20:59, 26.13it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 761/33666 [00:29<20:59, 26.13it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 761/33666 [00:29<20:59, 26.13it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 762/33666 [00:29<20:59, 26.13it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 762/33666 [00:29<20:59, 26.13it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 763/33666 [00:29<20:59, 26.12it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 763/33666 [00:29<20:59, 26.12it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 764/33666 [00:29<20:59, 26.13it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 764/33666 [00:29<20:59, 26.13it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 765/33666 [00:29<20:58, 26.14it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 765/33666 [00:29<20:58, 26.14it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 766/33666 [00:29<20:58, 26.13it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 766/33666 [00:29<20:58, 26.13it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 767/33666 [00:29<20:58, 26.14it/s, loss=4.59, v_num=0]
Epoch 0:   2%|▏         | 767/33666 [00:29<20:58, 26.14it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 768/33666 [00:29<20:58, 26.14it/s, loss=4.6, v_num=0]
Epoch 0:   2%|▏         | 768/33666 [00:29<20:58, 26.14it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 769/33666 [00:29<20:58, 26.14it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 769/33666 [00:29<20:58, 26.14it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 770/33666 [00:29<20:58, 26.14it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 770/33666 [00:29<20:58, 26.14it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 771/33666 [00:29<20:58, 26.14it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 771/33666 [00:29<20:58, 26.14it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 772/33666 [00:29<20:58, 26.15it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 772/33666 [00:29<20:58, 26.15it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 773/33666 [00:29<20:57, 26.15it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 773/33666 [00:29<20:57, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 774/33666 [00:29<20:57, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 774/33666 [00:29<20:57, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 775/33666 [00:29<20:57, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 775/33666 [00:29<20:57, 26.16it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 776/33666 [00:29<20:56, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 776/33666 [00:29<20:56, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 777/33666 [00:29<20:56, 26.18it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 777/33666 [00:29<20:56, 26.18it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 778/33666 [00:29<20:56, 26.18it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 778/33666 [00:29<20:56, 26.18it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 779/33666 [00:29<20:55, 26.19it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 779/33666 [00:29<20:55, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 780/33666 [00:29<20:55, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 780/33666 [00:29<20:55, 26.19it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 781/33666 [00:29<20:55, 26.20it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 781/33666 [00:29<20:55, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 782/33666 [00:29<20:54, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 782/33666 [00:29<20:54, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 783/33666 [00:29<20:54, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 783/33666 [00:29<20:54, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 784/33666 [00:29<20:54, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 784/33666 [00:29<20:54, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 785/33666 [00:29<20:54, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 785/33666 [00:29<20:54, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 786/33666 [00:29<20:54, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 786/33666 [00:29<20:54, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 787/33666 [00:30<20:54, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 787/33666 [00:30<20:54, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 788/33666 [00:30<20:54, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 788/33666 [00:30<20:54, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 789/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 789/33666 [00:30<20:54, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 790/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 790/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 791/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 791/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 792/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 792/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 793/33666 [00:30<20:53, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 793/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 794/33666 [00:30<20:53, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 794/33666 [00:30<20:53, 26.23it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 795/33666 [00:30<20:53, 26.23it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 795/33666 [00:30<20:53, 26.23it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 796/33666 [00:30<20:53, 26.23it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 796/33666 [00:30<20:53, 26.23it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 797/33666 [00:30<20:52, 26.23it/s, loss=4.62, v_num=0]
Epoch 0:   2%|▏         | 797/33666 [00:30<20:53, 26.23it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 798/33666 [00:30<20:53, 26.23it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 798/33666 [00:30<20:53, 26.23it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 799/33666 [00:30<20:53, 26.23it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 799/33666 [00:30<20:53, 26.22it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 800/33666 [00:30<20:53, 26.22it/s, loss=4.61, v_num=0]
Epoch 0:   2%|▏         | 800/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 801/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 801/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 802/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 802/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 803/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 803/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 804/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 804/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 805/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 805/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 806/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 806/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 807/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 807/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 808/33666 [00:30<20:53, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 808/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 809/33666 [00:30<20:53, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 809/33666 [00:30<20:53, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 810/33666 [00:30<20:53, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 810/33666 [00:30<20:53, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 811/33666 [00:30<20:52, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 811/33666 [00:30<20:53, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 812/33666 [00:30<20:52, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 812/33666 [00:30<20:52, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 813/33666 [00:31<20:52, 26.23it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 813/33666 [00:31<20:52, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 814/33666 [00:31<20:52, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 814/33666 [00:31<20:52, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 815/33666 [00:31<20:52, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 815/33666 [00:31<20:52, 26.23it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 816/33666 [00:31<20:52, 26.24it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 816/33666 [00:31<20:52, 26.24it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 817/33666 [00:31<20:51, 26.24it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 817/33666 [00:31<20:51, 26.24it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 818/33666 [00:31<20:52, 26.23it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 818/33666 [00:31<20:52, 26.23it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 819/33666 [00:31<20:52, 26.23it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 819/33666 [00:31<20:52, 26.23it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 820/33666 [00:31<20:52, 26.23it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 820/33666 [00:31<20:52, 26.23it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 821/33666 [00:31<20:52, 26.23it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 821/33666 [00:31<20:52, 26.23it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 822/33666 [00:31<20:52, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 822/33666 [00:31<20:52, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 823/33666 [00:31<20:52, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 823/33666 [00:31<20:52, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 824/33666 [00:31<20:52, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 824/33666 [00:31<20:52, 26.22it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 825/33666 [00:31<20:52, 26.22it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 825/33666 [00:31<20:52, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 826/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 826/33666 [00:31<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 827/33666 [00:31<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 827/33666 [00:31<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 828/33666 [00:31<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 828/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 829/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 829/33666 [00:31<20:52, 26.21it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 830/33666 [00:31<20:52, 26.22it/s, loss=4.63, v_num=0]
Epoch 0:   2%|▏         | 830/33666 [00:31<20:52, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 831/33666 [00:31<20:52, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   2%|▏         | 831/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 832/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 832/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 833/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 833/33666 [00:31<20:52, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 834/33666 [00:31<20:52, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 834/33666 [00:31<20:52, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 835/33666 [00:31<20:52, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 835/33666 [00:31<20:52, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 836/33666 [00:31<20:52, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   2%|▏         | 836/33666 [00:31<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 837/33666 [00:31<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 837/33666 [00:31<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 838/33666 [00:31<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 838/33666 [00:31<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 839/33666 [00:32<20:52, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   2%|▏         | 839/33666 [00:32<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 840/33666 [00:32<20:52, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   2%|▏         | 840/33666 [00:32<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 841/33666 [00:32<20:52, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   2%|▏         | 841/33666 [00:32<20:52, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 842/33666 [00:32<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 842/33666 [00:32<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 843/33666 [00:32<20:52, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 843/33666 [00:32<20:52, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 844/33666 [00:32<20:52, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 844/33666 [00:32<20:52, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 845/33666 [00:32<20:52, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 845/33666 [00:32<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 846/33666 [00:32<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 846/33666 [00:32<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 847/33666 [00:32<20:52, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 847/33666 [00:32<20:52, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 848/33666 [00:32<20:52, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 848/33666 [00:32<20:52, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 849/33666 [00:32<20:52, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 849/33666 [00:32<20:52, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 850/33666 [00:32<20:52, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 850/33666 [00:32<20:52, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 851/33666 [00:32<20:52, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 851/33666 [00:32<20:52, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 852/33666 [00:32<20:52, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 852/33666 [00:32<20:52, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 853/33666 [00:32<20:52, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 853/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 854/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 854/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 855/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 855/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 856/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 856/33666 [00:32<20:52, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 857/33666 [00:32<20:51, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 857/33666 [00:32<20:52, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 858/33666 [00:32<20:51, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 858/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 859/33666 [00:32<20:51, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 859/33666 [00:32<20:51, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 860/33666 [00:32<20:52, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 860/33666 [00:32<20:52, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 861/33666 [00:32<20:52, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 861/33666 [00:32<20:52, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 862/33666 [00:32<20:51, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 862/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 863/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 863/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 864/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 864/33666 [00:32<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 865/33666 [00:33<20:52, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 865/33666 [00:33<20:52, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 866/33666 [00:33<20:52, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 866/33666 [00:33<20:52, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 867/33666 [00:33<20:52, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 867/33666 [00:33<20:52, 26.19it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 868/33666 [00:33<20:52, 26.19it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 868/33666 [00:33<20:52, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 869/33666 [00:33<20:52, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 869/33666 [00:33<20:52, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 870/33666 [00:33<20:52, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 870/33666 [00:33<20:52, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 871/33666 [00:33<20:52, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 871/33666 [00:33<20:52, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 872/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 872/33666 [00:33<20:52, 26.18it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 873/33666 [00:33<20:52, 26.18it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 873/33666 [00:33<20:52, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 874/33666 [00:33<20:52, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 874/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 875/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 875/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 876/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 876/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 877/33666 [00:33<20:52, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 877/33666 [00:33<20:52, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 878/33666 [00:33<20:52, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 878/33666 [00:33<20:52, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 879/33666 [00:33<20:52, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 879/33666 [00:33<20:52, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 880/33666 [00:33<20:52, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 880/33666 [00:33<20:52, 26.17it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 881/33666 [00:33<20:52, 26.17it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 881/33666 [00:33<20:52, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 882/33666 [00:33<20:52, 26.18it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 882/33666 [00:33<20:52, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 883/33666 [00:33<20:52, 26.18it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 883/33666 [00:33<20:52, 26.18it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 884/33666 [00:33<20:52, 26.18it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 884/33666 [00:33<20:52, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 885/33666 [00:33<20:52, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 885/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 886/33666 [00:33<20:52, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 886/33666 [00:33<20:52, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 887/33666 [00:33<20:52, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 887/33666 [00:33<20:52, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 888/33666 [00:33<20:52, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 888/33666 [00:33<20:52, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 889/33666 [00:33<20:52, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 889/33666 [00:33<20:52, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 890/33666 [00:33<20:52, 26.18it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 890/33666 [00:33<20:52, 26.18it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 891/33666 [00:34<20:52, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 891/33666 [00:34<20:52, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 892/33666 [00:34<20:52, 26.18it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 892/33666 [00:34<20:52, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 893/33666 [00:34<20:52, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 893/33666 [00:34<20:52, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 894/33666 [00:34<20:52, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 894/33666 [00:34<20:52, 26.17it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 895/33666 [00:34<20:52, 26.17it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 895/33666 [00:34<20:52, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 896/33666 [00:34<20:51, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 896/33666 [00:34<20:51, 26.17it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 897/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 897/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 898/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 898/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 899/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 899/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 900/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 900/33666 [00:34<20:51, 26.18it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 901/33666 [00:34<20:51, 26.18it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 901/33666 [00:34<20:51, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 902/33666 [00:34<20:51, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 902/33666 [00:34<20:51, 26.18it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 903/33666 [00:34<20:51, 26.18it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 903/33666 [00:34<20:51, 26.18it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 904/33666 [00:34<20:51, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 904/33666 [00:34<20:51, 26.18it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 905/33666 [00:34<20:51, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 905/33666 [00:34<20:51, 26.19it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 906/33666 [00:34<20:51, 26.18it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 906/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 907/33666 [00:34<20:51, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 907/33666 [00:34<20:51, 26.18it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 908/33666 [00:34<20:50, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 908/33666 [00:34<20:50, 26.19it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 909/33666 [00:34<20:50, 26.19it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 909/33666 [00:34<20:50, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 910/33666 [00:34<20:50, 26.19it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 910/33666 [00:34<20:50, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 911/33666 [00:34<20:50, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 911/33666 [00:34<20:50, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 912/33666 [00:34<20:50, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 912/33666 [00:34<20:50, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 913/33666 [00:34<20:50, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 913/33666 [00:34<20:50, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 914/33666 [00:34<20:50, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 914/33666 [00:34<20:50, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 915/33666 [00:34<20:50, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 915/33666 [00:34<20:50, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 916/33666 [00:34<20:50, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 916/33666 [00:34<20:50, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 917/33666 [00:35<20:50, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 917/33666 [00:35<20:50, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 918/33666 [00:35<20:50, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 918/33666 [00:35<20:50, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 919/33666 [00:35<20:50, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 919/33666 [00:35<20:50, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 920/33666 [00:35<20:50, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 920/33666 [00:35<20:50, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 921/33666 [00:35<20:49, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 921/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 922/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 922/33666 [00:35<20:49, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 923/33666 [00:35<20:50, 26.19it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 923/33666 [00:35<20:50, 26.19it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 924/33666 [00:35<20:50, 26.19it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 924/33666 [00:35<20:50, 26.19it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 925/33666 [00:35<20:49, 26.19it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 925/33666 [00:35<20:49, 26.19it/s, loss=4.57, v_num=0]
Epoch 0:   3%|▎         | 926/33666 [00:35<20:49, 26.20it/s, loss=4.57, v_num=0]
Epoch 0:   3%|▎         | 926/33666 [00:35<20:49, 26.20it/s, loss=4.56, v_num=0]
Epoch 0:   3%|▎         | 927/33666 [00:35<20:49, 26.20it/s, loss=4.56, v_num=0]
Epoch 0:   3%|▎         | 927/33666 [00:35<20:49, 26.20it/s, loss=4.57, v_num=0]
Epoch 0:   3%|▎         | 928/33666 [00:35<20:49, 26.20it/s, loss=4.57, v_num=0]
Epoch 0:   3%|▎         | 928/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 929/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 929/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 930/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 930/33666 [00:35<20:49, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 931/33666 [00:35<20:49, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 931/33666 [00:35<20:49, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 932/33666 [00:35<20:49, 26.20it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 932/33666 [00:35<20:49, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 933/33666 [00:35<20:49, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 933/33666 [00:35<20:49, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 934/33666 [00:35<20:49, 26.20it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 934/33666 [00:35<20:49, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 935/33666 [00:35<20:49, 26.20it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 935/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 936/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 936/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 937/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 937/33666 [00:35<20:49, 26.20it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 938/33666 [00:35<20:48, 26.21it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 938/33666 [00:35<20:48, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 939/33666 [00:35<20:48, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 939/33666 [00:35<20:48, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 940/33666 [00:35<20:48, 26.22it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 940/33666 [00:35<20:48, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 941/33666 [00:35<20:48, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 941/33666 [00:35<20:48, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 942/33666 [00:35<20:47, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 942/33666 [00:35<20:47, 26.22it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 943/33666 [00:35<20:47, 26.23it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 943/33666 [00:35<20:47, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 944/33666 [00:35<20:47, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 944/33666 [00:35<20:47, 26.23it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 945/33666 [00:36<20:47, 26.23it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 945/33666 [00:36<20:47, 26.23it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 946/33666 [00:36<20:47, 26.23it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 946/33666 [00:36<20:47, 26.23it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 947/33666 [00:36<20:47, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 947/33666 [00:36<20:47, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 948/33666 [00:36<20:47, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 948/33666 [00:36<20:47, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 949/33666 [00:36<20:47, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 949/33666 [00:36<20:47, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 950/33666 [00:36<20:47, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 950/33666 [00:36<20:47, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 951/33666 [00:36<20:47, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 951/33666 [00:36<20:48, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 952/33666 [00:36<20:48, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 952/33666 [00:36<20:48, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 953/33666 [00:36<20:47, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 953/33666 [00:36<20:48, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 954/33666 [00:36<20:47, 26.22it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 954/33666 [00:36<20:47, 26.22it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 955/33666 [00:36<20:47, 26.22it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 955/33666 [00:36<20:47, 26.22it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 956/33666 [00:36<20:47, 26.21it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 956/33666 [00:36<20:47, 26.21it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 957/33666 [00:36<20:47, 26.21it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 957/33666 [00:36<20:47, 26.21it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 958/33666 [00:36<20:47, 26.21it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 958/33666 [00:36<20:47, 26.21it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 959/33666 [00:36<20:47, 26.21it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 959/33666 [00:36<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 960/33666 [00:36<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 960/33666 [00:36<20:47, 26.21it/s, loss=4.57, v_num=0]
Epoch 0:   3%|▎         | 961/33666 [00:36<20:47, 26.21it/s, loss=4.57, v_num=0]
Epoch 0:   3%|▎         | 961/33666 [00:36<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 962/33666 [00:36<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 962/33666 [00:36<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 963/33666 [00:36<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 963/33666 [00:36<20:47, 26.21it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 964/33666 [00:36<20:47, 26.21it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 964/33666 [00:36<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 965/33666 [00:36<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 965/33666 [00:36<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 966/33666 [00:36<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 966/33666 [00:36<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 967/33666 [00:36<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 967/33666 [00:36<20:47, 26.21it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 968/33666 [00:36<20:47, 26.21it/s, loss=4.58, v_num=0]
Epoch 0:   3%|▎         | 968/33666 [00:36<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 969/33666 [00:36<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 969/33666 [00:36<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 970/33666 [00:37<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 970/33666 [00:37<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 971/33666 [00:37<20:47, 26.21it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 971/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 972/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 972/33666 [00:37<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 973/33666 [00:37<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 973/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 974/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 974/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 975/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 975/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 976/33666 [00:37<20:47, 26.21it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 976/33666 [00:37<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 977/33666 [00:37<20:47, 26.21it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 977/33666 [00:37<20:47, 26.21it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 978/33666 [00:37<20:47, 26.21it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 978/33666 [00:37<20:47, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 979/33666 [00:37<20:47, 26.21it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 979/33666 [00:37<20:47, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 980/33666 [00:37<20:47, 26.20it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 980/33666 [00:37<20:47, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 981/33666 [00:37<20:47, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 981/33666 [00:37<20:47, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 982/33666 [00:37<20:47, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 982/33666 [00:37<20:47, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 983/33666 [00:37<20:47, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 983/33666 [00:37<20:47, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 984/33666 [00:37<20:47, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 984/33666 [00:37<20:47, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 985/33666 [00:37<20:47, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 985/33666 [00:37<20:47, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 986/33666 [00:37<20:47, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 986/33666 [00:37<20:47, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 987/33666 [00:37<20:47, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 987/33666 [00:37<20:47, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 988/33666 [00:37<20:47, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 988/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 989/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 989/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 990/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 990/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 991/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 991/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 992/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 992/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 993/33666 [00:37<20:47, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 993/33666 [00:37<20:47, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 994/33666 [00:37<20:47, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 994/33666 [00:37<20:47, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 995/33666 [00:37<20:47, 26.19it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 995/33666 [00:37<20:47, 26.19it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 996/33666 [00:38<20:47, 26.19it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 996/33666 [00:38<20:47, 26.19it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 997/33666 [00:38<20:47, 26.19it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 997/33666 [00:38<20:47, 26.19it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 998/33666 [00:38<20:47, 26.19it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 998/33666 [00:38<20:47, 26.19it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 999/33666 [00:38<20:47, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 999/33666 [00:38<20:47, 26.18it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1000/33666 [00:38<20:47, 26.18it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1000/33666 [00:38<20:47, 26.18it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1001/33666 [00:38<20:48, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1001/33666 [00:38<20:48, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1002/33666 [00:38<20:48, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1002/33666 [00:38<20:48, 26.17it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1003/33666 [00:38<20:48, 26.17it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1003/33666 [00:38<20:48, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1004/33666 [00:38<20:48, 26.17it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1004/33666 [00:38<20:48, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1005/33666 [00:38<20:48, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1005/33666 [00:38<20:48, 26.16it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1006/33666 [00:38<20:48, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1006/33666 [00:38<20:48, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1007/33666 [00:38<20:48, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1007/33666 [00:38<20:48, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1008/33666 [00:38<20:48, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1008/33666 [00:38<20:48, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1009/33666 [00:38<20:48, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1009/33666 [00:38<20:48, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1010/33666 [00:38<20:48, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1010/33666 [00:38<20:48, 26.15it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1011/33666 [00:38<20:48, 26.15it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1011/33666 [00:38<20:48, 26.15it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1012/33666 [00:38<20:48, 26.15it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1012/33666 [00:38<20:48, 26.15it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1013/33666 [00:38<20:48, 26.15it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1013/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1014/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1014/33666 [00:38<20:48, 26.15it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1015/33666 [00:38<20:48, 26.15it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1015/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1016/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1016/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1017/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1017/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1018/33666 [00:38<20:48, 26.15it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1018/33666 [00:38<20:48, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1019/33666 [00:38<20:48, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1019/33666 [00:38<20:48, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1020/33666 [00:39<20:48, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1020/33666 [00:39<20:48, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1021/33666 [00:39<20:48, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1021/33666 [00:39<20:48, 26.15it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1022/33666 [00:39<20:48, 26.16it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1022/33666 [00:39<20:48, 26.15it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1023/33666 [00:39<20:48, 26.15it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1023/33666 [00:39<20:48, 26.15it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1024/33666 [00:39<20:48, 26.15it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1024/33666 [00:39<20:48, 26.15it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1025/33666 [00:39<20:48, 26.15it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1025/33666 [00:39<20:48, 26.15it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1026/33666 [00:39<20:47, 26.15it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1026/33666 [00:39<20:47, 26.15it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1027/33666 [00:39<20:47, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1027/33666 [00:39<20:47, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1028/33666 [00:39<20:47, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1028/33666 [00:39<20:47, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1029/33666 [00:39<20:47, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1029/33666 [00:39<20:47, 26.16it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1030/33666 [00:39<20:47, 26.16it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1030/33666 [00:39<20:47, 26.16it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1031/33666 [00:39<20:47, 26.15it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1031/33666 [00:39<20:47, 26.15it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1032/33666 [00:39<20:47, 26.15it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1032/33666 [00:39<20:47, 26.15it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1033/33666 [00:39<20:47, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1033/33666 [00:39<20:47, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1034/33666 [00:39<20:47, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1034/33666 [00:39<20:47, 26.15it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1035/33666 [00:39<20:47, 26.16it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1035/33666 [00:39<20:47, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1036/33666 [00:39<20:47, 26.16it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1036/33666 [00:39<20:47, 26.16it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1037/33666 [00:39<20:47, 26.16it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1037/33666 [00:39<20:47, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1038/33666 [00:39<20:47, 26.16it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1038/33666 [00:39<20:47, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1039/33666 [00:39<20:47, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1039/33666 [00:39<20:47, 26.16it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1040/33666 [00:39<20:46, 26.16it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1040/33666 [00:39<20:46, 26.16it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1041/33666 [00:39<20:46, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1041/33666 [00:39<20:46, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1042/33666 [00:39<20:46, 26.17it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1042/33666 [00:39<20:46, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1043/33666 [00:39<20:46, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1043/33666 [00:39<20:46, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1044/33666 [00:39<20:46, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1044/33666 [00:39<20:46, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1045/33666 [00:39<20:46, 26.17it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1045/33666 [00:39<20:46, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1046/33666 [00:39<20:46, 26.17it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1046/33666 [00:39<20:46, 26.17it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1047/33666 [00:39<20:46, 26.18it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1047/33666 [00:39<20:46, 26.18it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1048/33666 [00:40<20:46, 26.18it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1048/33666 [00:40<20:46, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1049/33666 [00:40<20:46, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1049/33666 [00:40<20:46, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1050/33666 [00:40<20:45, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1050/33666 [00:40<20:45, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1051/33666 [00:40<20:45, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1051/33666 [00:40<20:45, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1052/33666 [00:40<20:45, 26.18it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1052/33666 [00:40<20:45, 26.18it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1053/33666 [00:40<20:45, 26.18it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1053/33666 [00:40<20:45, 26.18it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1054/33666 [00:40<20:45, 26.19it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1054/33666 [00:40<20:45, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1055/33666 [00:40<20:45, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1055/33666 [00:40<20:45, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1056/33666 [00:40<20:45, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1056/33666 [00:40<20:45, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1057/33666 [00:40<20:45, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1057/33666 [00:40<20:45, 26.19it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1058/33666 [00:40<20:44, 26.19it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1058/33666 [00:40<20:44, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1059/33666 [00:40<20:44, 26.19it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1059/33666 [00:40<20:44, 26.19it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1060/33666 [00:40<20:44, 26.19it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1060/33666 [00:40<20:44, 26.19it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1061/33666 [00:40<20:44, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1061/33666 [00:40<20:44, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1062/33666 [00:40<20:44, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1062/33666 [00:40<20:44, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1063/33666 [00:40<20:44, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1063/33666 [00:40<20:44, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1064/33666 [00:40<20:44, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1064/33666 [00:40<20:44, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1065/33666 [00:40<20:44, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1065/33666 [00:40<20:44, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1066/33666 [00:40<20:44, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1066/33666 [00:40<20:44, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1067/33666 [00:40<20:44, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1067/33666 [00:40<20:44, 26.20it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1068/33666 [00:40<20:43, 26.21it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1068/33666 [00:40<20:43, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1069/33666 [00:40<20:43, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1069/33666 [00:40<20:43, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1070/33666 [00:40<20:43, 26.20it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1070/33666 [00:40<20:43, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1071/33666 [00:40<20:43, 26.21it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1071/33666 [00:40<20:43, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1072/33666 [00:40<20:43, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1072/33666 [00:40<20:43, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1073/33666 [00:40<20:43, 26.20it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1073/33666 [00:40<20:43, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1074/33666 [00:40<20:43, 26.20it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1074/33666 [00:40<20:43, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1075/33666 [00:41<20:43, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1075/33666 [00:41<20:43, 26.20it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1076/33666 [00:41<20:43, 26.21it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1076/33666 [00:41<20:43, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1077/33666 [00:41<20:43, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1077/33666 [00:41<20:43, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1078/33666 [00:41<20:43, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1078/33666 [00:41<20:43, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1079/33666 [00:41<20:43, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1079/33666 [00:41<20:43, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1080/33666 [00:41<20:43, 26.21it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1080/33666 [00:41<20:43, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1081/33666 [00:41<20:43, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1081/33666 [00:41<20:43, 26.21it/s, loss=4.7, v_num=0]
Epoch 0:   3%|▎         | 1082/33666 [00:41<20:42, 26.21it/s, loss=4.7, v_num=0]
Epoch 0:   3%|▎         | 1082/33666 [00:41<20:42, 26.21it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1083/33666 [00:41<20:42, 26.22it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1083/33666 [00:41<20:42, 26.22it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1084/33666 [00:41<20:42, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1084/33666 [00:41<20:42, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1085/33666 [00:41<20:42, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1085/33666 [00:41<20:42, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1086/33666 [00:41<20:42, 26.21it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1086/33666 [00:41<20:42, 26.21it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1087/33666 [00:41<20:42, 26.22it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1087/33666 [00:41<20:42, 26.22it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1088/33666 [00:41<20:42, 26.22it/s, loss=4.69, v_num=0]
Epoch 0:   3%|▎         | 1088/33666 [00:41<20:42, 26.22it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1089/33666 [00:41<20:42, 26.22it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1089/33666 [00:41<20:42, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1090/33666 [00:41<20:42, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1090/33666 [00:41<20:42, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1091/33666 [00:41<20:42, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1091/33666 [00:41<20:42, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1092/33666 [00:41<20:42, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1092/33666 [00:41<20:42, 26.22it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1093/33666 [00:41<20:42, 26.22it/s, loss=4.68, v_num=0]
Epoch 0:   3%|▎         | 1093/33666 [00:41<20:42, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1094/33666 [00:41<20:42, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1094/33666 [00:41<20:42, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1095/33666 [00:41<20:41, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1095/33666 [00:41<20:41, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1096/33666 [00:41<20:42, 26.22it/s, loss=4.67, v_num=0]
Epoch 0:   3%|▎         | 1096/33666 [00:41<20:42, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1097/33666 [00:41<20:41, 26.22it/s, loss=4.66, v_num=0]
Epoch 0:   3%|▎         | 1097/33666 [00:41<20:41, 26.22it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1098/33666 [00:41<20:41, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1098/33666 [00:41<20:41, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1099/33666 [00:41<20:41, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1099/33666 [00:41<20:41, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1100/33666 [00:41<20:41, 26.23it/s, loss=4.65, v_num=0]
Epoch 0:   3%|▎         | 1100/33666 [00:41<20:41, 26.23it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1101/33666 [00:41<20:41, 26.23it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1101/33666 [00:41<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1102/33666 [00:42<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1102/33666 [00:42<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1103/33666 [00:42<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1103/33666 [00:42<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1104/33666 [00:42<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1104/33666 [00:42<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1105/33666 [00:42<20:41, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1105/33666 [00:42<20:41, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1106/33666 [00:42<20:41, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1106/33666 [00:42<20:41, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1107/33666 [00:42<20:41, 26.23it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1107/33666 [00:42<20:41, 26.23it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1108/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1108/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1109/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1109/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1110/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1110/33666 [00:42<20:40, 26.24it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1111/33666 [00:42<20:40, 26.24it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1111/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1112/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1112/33666 [00:42<20:40, 26.24it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1113/33666 [00:42<20:40, 26.24it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1113/33666 [00:42<20:40, 26.24it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1114/33666 [00:42<20:40, 26.24it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1114/33666 [00:42<20:40, 26.24it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1115/33666 [00:42<20:40, 26.24it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1115/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1116/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1116/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1117/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1117/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1118/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1118/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1119/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1119/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1120/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1120/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1121/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1121/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1122/33666 [00:42<20:40, 26.24it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1122/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1123/33666 [00:42<20:39, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1123/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1124/33666 [00:42<20:39, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1124/33666 [00:42<20:40, 26.24it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1125/33666 [00:42<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1125/33666 [00:42<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1126/33666 [00:42<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1126/33666 [00:42<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1127/33666 [00:42<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1127/33666 [00:42<20:39, 26.25it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1128/33666 [00:42<20:39, 26.25it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1128/33666 [00:42<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1129/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1129/33666 [00:43<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1130/33666 [00:43<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1130/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1131/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1131/33666 [00:43<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1132/33666 [00:43<20:39, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1132/33666 [00:43<20:39, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1133/33666 [00:43<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1133/33666 [00:43<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1134/33666 [00:43<20:39, 26.25it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1134/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1135/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1135/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1136/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1136/33666 [00:43<20:39, 26.25it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1137/33666 [00:43<20:38, 26.26it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1137/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1138/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1138/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1139/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1139/33666 [00:43<20:38, 26.26it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1140/33666 [00:43<20:38, 26.26it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1140/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1141/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1141/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1142/33666 [00:43<20:38, 26.26it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1142/33666 [00:43<20:38, 26.26it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1143/33666 [00:43<20:38, 26.26it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1143/33666 [00:43<20:38, 26.26it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1144/33666 [00:43<20:38, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1144/33666 [00:43<20:38, 26.27it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1145/33666 [00:43<20:38, 26.27it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1145/33666 [00:43<20:38, 26.27it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1146/33666 [00:43<20:38, 26.27it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1146/33666 [00:43<20:38, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1147/33666 [00:43<20:38, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1147/33666 [00:43<20:38, 26.26it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1148/33666 [00:43<20:38, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1148/33666 [00:43<20:38, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1149/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1149/33666 [00:43<20:37, 26.27it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1150/33666 [00:43<20:37, 26.27it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1150/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1151/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1151/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1152/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1152/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1153/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1153/33666 [00:43<20:37, 26.27it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1154/33666 [00:43<20:37, 26.28it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1154/33666 [00:43<20:37, 26.28it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1155/33666 [00:43<20:37, 26.28it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1155/33666 [00:43<20:37, 26.28it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1156/33666 [00:43<20:37, 26.28it/s, loss=4.64, v_num=0]
Epoch 0:   3%|▎         | 1156/33666 [00:43<20:37, 26.28it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1157/33666 [00:44<20:37, 26.28it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1157/33666 [00:44<20:37, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1158/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1158/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1159/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1159/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1160/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1160/33666 [00:44<20:36, 26.28it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1161/33666 [00:44<20:36, 26.28it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1161/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1162/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1162/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1163/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1163/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1164/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1164/33666 [00:44<20:36, 26.28it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1165/33666 [00:44<20:36, 26.28it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1165/33666 [00:44<20:36, 26.28it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1166/33666 [00:44<20:36, 26.29it/s, loss=4.63, v_num=0]
Epoch 0:   3%|▎         | 1166/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1167/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1167/33666 [00:44<20:36, 26.29it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1168/33666 [00:44<20:36, 26.28it/s, loss=4.62, v_num=0]
Epoch 0:   3%|▎         | 1168/33666 [00:44<20:36, 26.28it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1169/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1169/33666 [00:44<20:36, 26.28it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1170/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1170/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1171/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1171/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1172/33666 [00:44<20:36, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1172/33666 [00:44<20:36, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1173/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1173/33666 [00:44<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 1174/33666 [00:44<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   3%|▎         | 1174/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1175/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1175/33666 [00:44<20:35, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1176/33666 [00:44<20:35, 26.29it/s, loss=4.61, v_num=0]
Epoch 0:   3%|▎         | 1176/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1177/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1177/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1178/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   3%|▎         | 1178/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▎         | 1179/33666 [00:44<20:35, 26.29it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▎         | 1179/33666 [00:44<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1180/33666 [00:44<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1180/33666 [00:44<20:35, 26.29it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▎         | 1181/33666 [00:44<20:35, 26.29it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▎         | 1181/33666 [00:44<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1182/33666 [00:44<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1182/33666 [00:44<20:35, 26.29it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▎         | 1183/33666 [00:44<20:35, 26.29it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▎         | 1183/33666 [00:44<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1184/33666 [00:45<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1184/33666 [00:45<20:35, 26.29it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1185/33666 [00:45<20:35, 26.30it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▎         | 1185/33666 [00:45<20:35, 26.30it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▎         | 1186/33666 [00:45<20:35, 26.30it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▎         | 1186/33666 [00:45<20:35, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1187/33666 [00:45<20:35, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1187/33666 [00:45<20:35, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1188/33666 [00:45<20:35, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1188/33666 [00:45<20:35, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1189/33666 [00:45<20:34, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1189/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1190/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1190/33666 [00:45<20:34, 26.30it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1191/33666 [00:45<20:34, 26.30it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1191/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1192/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1192/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1193/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1193/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1194/33666 [00:45<20:34, 26.31it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1194/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1195/33666 [00:45<20:34, 26.30it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1195/33666 [00:45<20:34, 26.30it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1196/33666 [00:45<20:34, 26.30it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1196/33666 [00:45<20:34, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1197/33666 [00:45<20:34, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1197/33666 [00:45<20:34, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1198/33666 [00:45<20:34, 26.31it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1198/33666 [00:45<20:34, 26.30it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1199/33666 [00:45<20:34, 26.31it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1199/33666 [00:45<20:34, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1200/33666 [00:45<20:34, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1200/33666 [00:45<20:34, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1201/33666 [00:45<20:33, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1201/33666 [00:45<20:34, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1202/33666 [00:45<20:33, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1202/33666 [00:45<20:33, 26.31it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1203/33666 [00:45<20:33, 26.31it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1203/33666 [00:45<20:33, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1204/33666 [00:45<20:33, 26.31it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1204/33666 [00:45<20:33, 26.31it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1205/33666 [00:45<20:33, 26.31it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1205/33666 [00:45<20:33, 26.31it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1206/33666 [00:45<20:33, 26.31it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1206/33666 [00:45<20:33, 26.31it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1207/33666 [00:45<20:33, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1207/33666 [00:45<20:33, 26.32it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1208/33666 [00:45<20:33, 26.32it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1208/33666 [00:45<20:33, 26.32it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1209/33666 [00:45<20:33, 26.32it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1209/33666 [00:45<20:33, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1210/33666 [00:45<20:33, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1210/33666 [00:45<20:33, 26.32it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▎         | 1211/33666 [00:46<20:33, 26.32it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▎         | 1211/33666 [00:46<20:33, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1212/33666 [00:46<20:32, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1212/33666 [00:46<20:33, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1213/33666 [00:46<20:32, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1213/33666 [00:46<20:32, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1214/33666 [00:46<20:32, 26.32it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▎         | 1214/33666 [00:46<20:32, 26.32it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1215/33666 [00:46<20:32, 26.32it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1215/33666 [00:46<20:32, 26.32it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1216/33666 [00:46<20:32, 26.33it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▎         | 1216/33666 [00:46<20:32, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1217/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1217/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1218/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1218/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1219/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1219/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1220/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1220/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1221/33666 [00:46<20:32, 26.32it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1221/33666 [00:46<20:32, 26.32it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1222/33666 [00:46<20:32, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1222/33666 [00:46<20:32, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1223/33666 [00:46<20:32, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1223/33666 [00:46<20:32, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1224/33666 [00:46<20:32, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1224/33666 [00:46<20:32, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1225/33666 [00:46<20:32, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1225/33666 [00:46<20:32, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1226/33666 [00:46<20:32, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1226/33666 [00:46<20:32, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1227/33666 [00:46<20:31, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1227/33666 [00:46<20:31, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1228/33666 [00:46<20:31, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1228/33666 [00:46<20:31, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1229/33666 [00:46<20:31, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1229/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1230/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1230/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1231/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1231/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1232/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1232/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1233/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1233/33666 [00:46<20:31, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1234/33666 [00:46<20:31, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1234/33666 [00:46<20:31, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1235/33666 [00:46<20:31, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1235/33666 [00:46<20:31, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1236/33666 [00:46<20:31, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1236/33666 [00:46<20:31, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1237/33666 [00:46<20:31, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1237/33666 [00:46<20:31, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1238/33666 [00:46<20:31, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1238/33666 [00:46<20:31, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1239/33666 [00:47<20:30, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1239/33666 [00:47<20:30, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1240/33666 [00:47<20:30, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1240/33666 [00:47<20:30, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1241/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1241/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1242/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1242/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1243/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1243/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1244/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1244/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1245/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1245/33666 [00:47<20:30, 26.35it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1246/33666 [00:47<20:30, 26.35it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1246/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1247/33666 [00:47<20:30, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1247/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1248/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1248/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1249/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1249/33666 [00:47<20:30, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1250/33666 [00:47<20:30, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1250/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1251/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1251/33666 [00:47<20:30, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1252/33666 [00:47<20:29, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1252/33666 [00:47<20:29, 26.35it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1253/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1253/33666 [00:47<20:29, 26.36it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1254/33666 [00:47<20:29, 26.36it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▎         | 1254/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1255/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1255/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1256/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1256/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1257/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1257/33666 [00:47<20:29, 26.36it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1258/33666 [00:47<20:29, 26.36it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▎         | 1258/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1259/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1259/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1260/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1260/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1261/33666 [00:47<20:29, 26.36it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▎         | 1261/33666 [00:47<20:29, 26.36it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1262/33666 [00:47<20:29, 26.37it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▎         | 1262/33666 [00:47<20:29, 26.37it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1263/33666 [00:47<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1263/33666 [00:47<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1264/33666 [00:47<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1264/33666 [00:47<20:29, 26.36it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1265/33666 [00:47<20:29, 26.36it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1265/33666 [00:47<20:29, 26.36it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1266/33666 [00:48<20:29, 26.36it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1266/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1267/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1267/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1268/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1268/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1269/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1269/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1270/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1270/33666 [00:48<20:29, 26.36it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1271/33666 [00:48<20:29, 26.35it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1271/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1272/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1272/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1273/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1273/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1274/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1274/33666 [00:48<20:29, 26.35it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1275/33666 [00:48<20:29, 26.35it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1275/33666 [00:48<20:29, 26.35it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1276/33666 [00:48<20:29, 26.35it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1276/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1277/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1277/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1278/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1278/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1279/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1279/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1280/33666 [00:48<20:29, 26.34it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1280/33666 [00:48<20:29, 26.34it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1281/33666 [00:48<20:29, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1281/33666 [00:48<20:29, 26.34it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1282/33666 [00:48<20:29, 26.34it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1282/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1283/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1283/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1284/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1284/33666 [00:48<20:29, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1285/33666 [00:48<20:29, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1285/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1286/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1286/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1287/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1287/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1288/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1288/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1289/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1289/33666 [00:48<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1290/33666 [00:48<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1290/33666 [00:48<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1291/33666 [00:49<20:29, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1291/33666 [00:49<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1292/33666 [00:49<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1292/33666 [00:49<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1293/33666 [00:49<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1293/33666 [00:49<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1294/33666 [00:49<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1294/33666 [00:49<20:29, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1295/33666 [00:49<20:29, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1295/33666 [00:49<20:29, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1296/33666 [00:49<20:29, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1296/33666 [00:49<20:29, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1297/33666 [00:49<20:29, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1297/33666 [00:49<20:29, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1298/33666 [00:49<20:28, 26.34it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1298/33666 [00:49<20:28, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1299/33666 [00:49<20:28, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1299/33666 [00:49<20:28, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1300/33666 [00:49<20:28, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1300/33666 [00:49<20:28, 26.34it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1301/33666 [00:49<20:28, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1301/33666 [00:49<20:29, 26.33it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1302/33666 [00:49<20:29, 26.33it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1302/33666 [00:49<20:29, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1303/33666 [00:49<20:29, 26.33it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1303/33666 [00:49<20:29, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1304/33666 [00:49<20:29, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1304/33666 [00:49<20:29, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1305/33666 [00:49<20:28, 26.33it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1305/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1306/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1306/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1307/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1307/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1308/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1308/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1309/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1309/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1310/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1310/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1311/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1311/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1312/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1312/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1313/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1313/33666 [00:49<20:28, 26.33it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1314/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1314/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1315/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1315/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1316/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1316/33666 [00:49<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1317/33666 [00:50<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1317/33666 [00:50<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1318/33666 [00:50<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1318/33666 [00:50<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1319/33666 [00:50<20:28, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1319/33666 [00:50<20:28, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1320/33666 [00:50<20:28, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1320/33666 [00:50<20:28, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1321/33666 [00:50<20:27, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1321/33666 [00:50<20:27, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1322/33666 [00:50<20:27, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1322/33666 [00:50<20:27, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1323/33666 [00:50<20:27, 26.34it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1323/33666 [00:50<20:27, 26.34it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1324/33666 [00:50<20:27, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1324/33666 [00:50<20:27, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1325/33666 [00:50<20:27, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1325/33666 [00:50<20:27, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1326/33666 [00:50<20:27, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1326/33666 [00:50<20:27, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1327/33666 [00:50<20:27, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1327/33666 [00:50<20:27, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1328/33666 [00:50<20:27, 26.35it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1328/33666 [00:50<20:27, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1329/33666 [00:50<20:27, 26.35it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1329/33666 [00:50<20:27, 26.35it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1330/33666 [00:50<20:27, 26.35it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1330/33666 [00:50<20:27, 26.35it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1331/33666 [00:50<20:26, 26.35it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1331/33666 [00:50<20:27, 26.35it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1332/33666 [00:50<20:26, 26.35it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1332/33666 [00:50<20:26, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1333/33666 [00:50<20:26, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1333/33666 [00:50<20:26, 26.35it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1334/33666 [00:50<20:26, 26.36it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1334/33666 [00:50<20:26, 26.36it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1335/33666 [00:50<20:26, 26.36it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1335/33666 [00:50<20:26, 26.36it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1336/33666 [00:50<20:26, 26.37it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1336/33666 [00:50<20:26, 26.36it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1337/33666 [00:50<20:26, 26.37it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1337/33666 [00:50<20:26, 26.37it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1338/33666 [00:50<20:25, 26.37it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1338/33666 [00:50<20:25, 26.37it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1339/33666 [00:50<20:25, 26.37it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1339/33666 [00:50<20:25, 26.37it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1340/33666 [00:50<20:25, 26.38it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1340/33666 [00:50<20:25, 26.38it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1341/33666 [00:50<20:25, 26.38it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1341/33666 [00:50<20:25, 26.38it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1342/33666 [00:50<20:25, 26.38it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1342/33666 [00:50<20:25, 26.38it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1343/33666 [00:50<20:24, 26.39it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1343/33666 [00:50<20:24, 26.39it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1344/33666 [00:50<20:24, 26.39it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1344/33666 [00:50<20:24, 26.39it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1345/33666 [00:50<20:24, 26.40it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1345/33666 [00:50<20:24, 26.39it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1346/33666 [00:50<20:24, 26.40it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1346/33666 [00:50<20:24, 26.40it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1347/33666 [00:51<20:24, 26.40it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1347/33666 [00:51<20:24, 26.40it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1348/33666 [00:51<20:23, 26.41it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1348/33666 [00:51<20:23, 26.41it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1349/33666 [00:51<20:23, 26.41it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1349/33666 [00:51<20:23, 26.41it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1350/33666 [00:51<20:23, 26.41it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1350/33666 [00:51<20:23, 26.41it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1351/33666 [00:51<20:23, 26.42it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1351/33666 [00:51<20:23, 26.42it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1352/33666 [00:51<20:23, 26.42it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1352/33666 [00:51<20:23, 26.42it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1353/33666 [00:51<20:22, 26.42it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1353/33666 [00:51<20:22, 26.42it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1354/33666 [00:51<20:22, 26.43it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1354/33666 [00:51<20:22, 26.43it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1355/33666 [00:51<20:22, 26.43it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1355/33666 [00:51<20:22, 26.43it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1356/33666 [00:51<20:22, 26.43it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1356/33666 [00:51<20:22, 26.43it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1357/33666 [00:51<20:22, 26.44it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1357/33666 [00:51<20:22, 26.44it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1358/33666 [00:51<20:21, 26.44it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1358/33666 [00:51<20:22, 26.44it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1359/33666 [00:51<20:21, 26.44it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1359/33666 [00:51<20:21, 26.44it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1360/33666 [00:51<20:21, 26.45it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1360/33666 [00:51<20:21, 26.45it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1361/33666 [00:51<20:21, 26.45it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1361/33666 [00:51<20:21, 26.45it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1362/33666 [00:51<20:21, 26.45it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1362/33666 [00:51<20:21, 26.45it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1363/33666 [00:51<20:21, 26.46it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1363/33666 [00:51<20:21, 26.46it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1364/33666 [00:51<20:20, 26.46it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1364/33666 [00:51<20:20, 26.46it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1365/33666 [00:51<20:20, 26.46it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1365/33666 [00:51<20:20, 26.46it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1366/33666 [00:51<20:20, 26.47it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1366/33666 [00:51<20:20, 26.47it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1367/33666 [00:51<20:20, 26.47it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1367/33666 [00:51<20:20, 26.47it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1368/33666 [00:51<20:20, 26.47it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1368/33666 [00:51<20:20, 26.47it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1369/33666 [00:51<20:19, 26.48it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1369/33666 [00:51<20:19, 26.48it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1370/33666 [00:51<20:19, 26.48it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1370/33666 [00:51<20:19, 26.48it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1371/33666 [00:51<20:19, 26.48it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1371/33666 [00:51<20:19, 26.48it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1372/33666 [00:51<20:19, 26.49it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1372/33666 [00:51<20:19, 26.49it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1373/33666 [00:51<20:19, 26.49it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1373/33666 [00:51<20:19, 26.49it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1374/33666 [00:51<20:18, 26.49it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1374/33666 [00:51<20:18, 26.49it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1375/33666 [00:51<20:18, 26.49it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1375/33666 [00:51<20:18, 26.49it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1376/33666 [00:51<20:18, 26.50it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1376/33666 [00:51<20:18, 26.50it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1377/33666 [00:51<20:18, 26.50it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1377/33666 [00:51<20:18, 26.50it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1378/33666 [00:51<20:18, 26.50it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1378/33666 [00:51<20:18, 26.50it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1379/33666 [00:52<20:18, 26.51it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1379/33666 [00:52<20:18, 26.51it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1380/33666 [00:52<20:17, 26.51it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1380/33666 [00:52<20:17, 26.51it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1381/33666 [00:52<20:17, 26.51it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1381/33666 [00:52<20:17, 26.51it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1382/33666 [00:52<20:17, 26.52it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1382/33666 [00:52<20:17, 26.52it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1383/33666 [00:52<20:17, 26.52it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1383/33666 [00:52<20:17, 26.52it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1384/33666 [00:52<20:17, 26.52it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1384/33666 [00:52<20:17, 26.52it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1385/33666 [00:52<20:16, 26.53it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1385/33666 [00:52<20:16, 26.53it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1386/33666 [00:52<20:16, 26.53it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1386/33666 [00:52<20:16, 26.53it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1387/33666 [00:52<20:16, 26.53it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1387/33666 [00:52<20:16, 26.53it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1388/33666 [00:52<20:16, 26.54it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1388/33666 [00:52<20:16, 26.54it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1389/33666 [00:52<20:16, 26.54it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1389/33666 [00:52<20:16, 26.54it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1390/33666 [00:52<20:15, 26.54it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1390/33666 [00:52<20:15, 26.54it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1391/33666 [00:52<20:15, 26.55it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1391/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1392/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1392/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1393/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1393/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1394/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1394/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1395/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1395/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1396/33666 [00:52<20:15, 26.55it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1396/33666 [00:52<20:15, 26.55it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1397/33666 [00:52<20:15, 26.55it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1397/33666 [00:52<20:15, 26.55it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1398/33666 [00:52<20:15, 26.55it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1398/33666 [00:52<20:15, 26.55it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1399/33666 [00:52<20:15, 26.55it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1399/33666 [00:52<20:15, 26.55it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1400/33666 [00:52<20:15, 26.55it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1400/33666 [00:52<20:15, 26.55it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1401/33666 [00:52<20:15, 26.55it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1401/33666 [00:52<20:15, 26.55it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1402/33666 [00:52<20:15, 26.55it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1402/33666 [00:52<20:15, 26.55it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1403/33666 [00:52<20:15, 26.55it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1403/33666 [00:52<20:15, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1404/33666 [00:52<20:15, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1404/33666 [00:52<20:15, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1405/33666 [00:52<20:15, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1405/33666 [00:52<20:15, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1406/33666 [00:52<20:15, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1406/33666 [00:52<20:15, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1407/33666 [00:52<20:14, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1407/33666 [00:52<20:14, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1408/33666 [00:53<20:14, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1408/33666 [00:53<20:14, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1409/33666 [00:53<20:14, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1409/33666 [00:53<20:14, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1410/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1410/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1411/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1411/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1412/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1412/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1413/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1413/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1414/33666 [00:53<20:14, 26.55it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1414/33666 [00:53<20:14, 26.55it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1415/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1415/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1416/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1416/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1417/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1417/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1418/33666 [00:53<20:14, 26.56it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1418/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1419/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1419/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1420/33666 [00:53<20:14, 26.56it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1420/33666 [00:53<20:14, 26.56it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1421/33666 [00:53<20:14, 26.56it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1421/33666 [00:53<20:14, 26.56it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1422/33666 [00:53<20:13, 26.56it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1422/33666 [00:53<20:13, 26.56it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1423/33666 [00:53<20:13, 26.56it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1423/33666 [00:53<20:13, 26.56it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1424/33666 [00:53<20:13, 26.56it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1424/33666 [00:53<20:13, 26.56it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1425/33666 [00:53<20:13, 26.56it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1425/33666 [00:53<20:13, 26.56it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1426/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1426/33666 [00:53<20:13, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1427/33666 [00:53<20:13, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1427/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1428/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1428/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1429/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1429/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1430/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1430/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1431/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1431/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1432/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1432/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1433/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1433/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1434/33666 [00:53<20:13, 26.57it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1434/33666 [00:53<20:13, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1435/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1435/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1436/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1436/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1437/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1437/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1438/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1438/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1439/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1439/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1440/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1440/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1441/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1441/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1442/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1442/33666 [00:54<20:12, 26.57it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1443/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1443/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1444/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1444/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1445/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1445/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1446/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1446/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1447/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1447/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1448/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1448/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1449/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1449/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1450/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1450/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1451/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1451/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1452/33666 [00:54<20:11, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1452/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1453/33666 [00:54<20:12, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1453/33666 [00:54<20:12, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1454/33666 [00:54<20:11, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1454/33666 [00:54<20:11, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1455/33666 [00:54<20:11, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1455/33666 [00:54<20:11, 26.58it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1456/33666 [00:54<20:11, 26.58it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1456/33666 [00:54<20:11, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1457/33666 [00:54<20:11, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1457/33666 [00:54<20:11, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1458/33666 [00:54<20:11, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1458/33666 [00:54<20:11, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1459/33666 [00:54<20:11, 26.58it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1459/33666 [00:54<20:11, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1460/33666 [00:54<20:11, 26.58it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1460/33666 [00:54<20:11, 26.58it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1461/33666 [00:54<20:11, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1461/33666 [00:54<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1462/33666 [00:54<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1462/33666 [00:54<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1463/33666 [00:55<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1463/33666 [00:55<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1464/33666 [00:55<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1464/33666 [00:55<20:11, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1465/33666 [00:55<20:11, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1465/33666 [00:55<20:11, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1466/33666 [00:55<20:11, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1466/33666 [00:55<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1467/33666 [00:55<20:11, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1467/33666 [00:55<20:11, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1468/33666 [00:55<20:11, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1468/33666 [00:55<20:11, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1469/33666 [00:55<20:11, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1469/33666 [00:55<20:11, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1470/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1470/33666 [00:55<20:10, 26.59it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1471/33666 [00:55<20:10, 26.59it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1471/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1472/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1472/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1473/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1473/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1474/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1474/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1475/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1475/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1476/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1476/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1477/33666 [00:55<20:10, 26.59it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1477/33666 [00:55<20:10, 26.59it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1478/33666 [00:55<20:10, 26.59it/s, loss=4.58, v_num=0]
Epoch 0:   4%|▍         | 1478/33666 [00:55<20:10, 26.59it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1479/33666 [00:55<20:10, 26.59it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1479/33666 [00:55<20:10, 26.59it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1480/33666 [00:55<20:10, 26.59it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1480/33666 [00:55<20:10, 26.59it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1481/33666 [00:55<20:10, 26.59it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1481/33666 [00:55<20:10, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1482/33666 [00:55<20:10, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1482/33666 [00:55<20:10, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1483/33666 [00:55<20:10, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1483/33666 [00:55<20:10, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1484/33666 [00:55<20:10, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1484/33666 [00:55<20:10, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1485/33666 [00:55<20:10, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1485/33666 [00:55<20:10, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1486/33666 [00:55<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1486/33666 [00:55<20:09, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1487/33666 [00:55<20:09, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1487/33666 [00:55<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1488/33666 [00:55<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1488/33666 [00:55<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1489/33666 [00:55<20:09, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1489/33666 [00:55<20:09, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1490/33666 [00:56<20:09, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1490/33666 [00:56<20:09, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1491/33666 [00:56<20:09, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1491/33666 [00:56<20:09, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1492/33666 [00:56<20:09, 26.59it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1492/33666 [00:56<20:09, 26.59it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1493/33666 [00:56<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1493/33666 [00:56<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1494/33666 [00:56<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1494/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1495/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1495/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1496/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1496/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1497/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1497/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1498/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1498/33666 [00:56<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1499/33666 [00:56<20:09, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1499/33666 [00:56<20:09, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1500/33666 [00:56<20:09, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1500/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1501/33666 [00:56<20:09, 26.60it/s, loss=4.62, v_num=0]
Epoch 0:   4%|▍         | 1501/33666 [00:56<20:09, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1502/33666 [00:56<20:09, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1502/33666 [00:56<20:09, 26.60it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1503/33666 [00:56<20:09, 26.60it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1503/33666 [00:56<20:09, 26.60it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1504/33666 [00:56<20:08, 26.60it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1504/33666 [00:56<20:08, 26.60it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1505/33666 [00:56<20:08, 26.60it/s, loss=4.59, v_num=0]
Epoch 0:   4%|▍         | 1505/33666 [00:56<20:08, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1506/33666 [00:56<20:08, 26.60it/s, loss=4.6, v_num=0]
Epoch 0:   4%|▍         | 1506/33666 [00:56<20:08, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1507/33666 [00:56<20:08, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1507/33666 [00:56<20:08, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1508/33666 [00:56<20:08, 26.60it/s, loss=4.61, v_num=0]
Epoch 0:   4%|▍         | 1508/33666 [00:56<20:08, 26.60it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1509/33666 [00:56<20:08, 26.61it/s, loss=4.63, v_num=0]
Epoch 0:   4%|▍         | 1509/33666 [00:56<20:08, 26.61it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1510/33666 [00:56<20:08, 26.61it/s, loss=4.64, v_num=0]
Epoch 0:   4%|▍         | 1510/33666 [00:56<20:08, 26.61it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1511/33666 [00:56<20:08, 26.61it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1511/33666 [00:56<20:08, 26.61it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1512/33666 [00:56<20:08, 26.61it/s, loss=4.65, v_num=0]
Epoch 0:   4%|▍         | 1512/33666 [00:56<20:08, 26.61it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1513/33666 [00:56<20:08, 26.62it/s, loss=4.66, v_num=0]
Epoch 0:   4%|▍         | 1513/33666 [00:56<20:08, 26.62it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1514/33666 [00:56<20:07, 26.62it/s, loss=4.67, v_num=0]
Epoch 0:   4%|▍         | 1514/33666 [00:56<20:07, 26.62it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1515/33666 [00:56<20:07, 26.62it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1515/33666 [00:56<20:07, 26.62it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1516/33666 [00:56<20:07, 26.62it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1516/33666 [00:56<20:07, 26.62it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1517/33666 [00:56<20:07, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1517/33666 [00:56<20:07, 26.63it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1518/33666 [00:57<20:07, 26.63it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1518/33666 [00:57<20:07, 26.63it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▍         | 1519/33666 [00:57<20:07, 26.63it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▍         | 1519/33666 [00:57<20:07, 26.63it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1520/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1520/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1521/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1521/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1522/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1522/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1523/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1523/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1524/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1524/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1525/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1525/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1526/33666 [00:57<20:06, 26.64it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1526/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1527/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1527/33666 [00:57<20:06, 26.64it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▍         | 1528/33666 [00:57<20:06, 26.64it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▍         | 1528/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1529/33666 [00:57<20:06, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1529/33666 [00:57<20:06, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1530/33666 [00:57<20:06, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1530/33666 [00:57<20:06, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1531/33666 [00:57<20:06, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1531/33666 [00:57<20:06, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1532/33666 [00:57<20:06, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1532/33666 [00:57<20:06, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1533/33666 [00:57<20:06, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1533/33666 [00:57<20:06, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1534/33666 [00:57<20:05, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1534/33666 [00:57<20:05, 26.64it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1535/33666 [00:57<20:05, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1535/33666 [00:57<20:05, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1536/33666 [00:57<20:05, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1536/33666 [00:57<20:05, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1537/33666 [00:57<20:05, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1537/33666 [00:57<20:05, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1538/33666 [00:57<20:05, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1538/33666 [00:57<20:05, 26.64it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1539/33666 [00:57<20:05, 26.64it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1539/33666 [00:57<20:05, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1540/33666 [00:57<20:05, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1540/33666 [00:57<20:05, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1541/33666 [00:57<20:05, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1541/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1542/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1542/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1543/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1543/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1544/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1544/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1545/33666 [00:57<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1545/33666 [00:57<20:05, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1546/33666 [00:58<20:05, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1546/33666 [00:58<20:05, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1547/33666 [00:58<20:05, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1547/33666 [00:58<20:05, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1548/33666 [00:58<20:05, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1548/33666 [00:58<20:05, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1549/33666 [00:58<20:05, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1549/33666 [00:58<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1550/33666 [00:58<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1550/33666 [00:58<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1551/33666 [00:58<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1551/33666 [00:58<20:05, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1552/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1552/33666 [00:58<20:04, 26.65it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1553/33666 [00:58<20:04, 26.65it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1553/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1554/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1554/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1555/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1555/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1556/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1556/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1557/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1557/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1558/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1558/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1559/33666 [00:58<20:04, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1559/33666 [00:58<20:04, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1560/33666 [00:58<20:04, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1560/33666 [00:58<20:04, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1561/33666 [00:58<20:04, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1561/33666 [00:58<20:04, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1562/33666 [00:58<20:04, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1562/33666 [00:58<20:04, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1563/33666 [00:58<20:04, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1563/33666 [00:58<20:04, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1564/33666 [00:58<20:04, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1564/33666 [00:58<20:04, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1565/33666 [00:58<20:04, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1565/33666 [00:58<20:04, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1566/33666 [00:58<20:04, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1566/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1567/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1567/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1568/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1568/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1569/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1569/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1570/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1570/33666 [00:58<20:04, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1571/33666 [00:58<20:04, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1571/33666 [00:58<20:04, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1572/33666 [00:58<20:03, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1572/33666 [00:58<20:03, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1573/33666 [00:59<20:03, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1573/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1574/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1574/33666 [00:59<20:03, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1575/33666 [00:59<20:03, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1575/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1576/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1576/33666 [00:59<20:03, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   5%|▍         | 1577/33666 [00:59<20:03, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   5%|▍         | 1577/33666 [00:59<20:03, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   5%|▍         | 1578/33666 [00:59<20:03, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   5%|▍         | 1578/33666 [00:59<20:03, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   5%|▍         | 1579/33666 [00:59<20:03, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   5%|▍         | 1579/33666 [00:59<20:03, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   5%|▍         | 1580/33666 [00:59<20:03, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   5%|▍         | 1580/33666 [00:59<20:03, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   5%|▍         | 1581/33666 [00:59<20:03, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   5%|▍         | 1581/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1582/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1582/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1583/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1583/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1584/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1584/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1585/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1585/33666 [00:59<20:03, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1586/33666 [00:59<20:03, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1586/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1587/33666 [00:59<20:03, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1587/33666 [00:59<20:03, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1588/33666 [00:59<20:02, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1588/33666 [00:59<20:02, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1589/33666 [00:59<20:02, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1589/33666 [00:59<20:02, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1590/33666 [00:59<20:02, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1590/33666 [00:59<20:02, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1591/33666 [00:59<20:02, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1591/33666 [00:59<20:02, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1592/33666 [00:59<20:02, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1592/33666 [00:59<20:02, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1593/33666 [00:59<20:02, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1593/33666 [00:59<20:02, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1594/33666 [00:59<20:02, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1594/33666 [00:59<20:02, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1595/33666 [00:59<20:02, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1595/33666 [00:59<20:02, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1596/33666 [00:59<20:02, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1596/33666 [00:59<20:02, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1597/33666 [00:59<20:02, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1597/33666 [00:59<20:02, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1598/33666 [00:59<20:02, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1598/33666 [00:59<20:02, 26.67it/s, loss=4.73, v_num=0]
Epoch 0:   5%|▍         | 1599/33666 [00:59<20:02, 26.67it/s, loss=4.73, v_num=0]
Epoch 0:   5%|▍         | 1599/33666 [00:59<20:02, 26.67it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1600/33666 [00:59<20:02, 26.67it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1600/33666 [00:59<20:02, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1601/33666 [01:00<20:02, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1601/33666 [01:00<20:02, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1602/33666 [01:00<20:02, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1602/33666 [01:00<20:02, 26.68it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1603/33666 [01:00<20:01, 26.68it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1603/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1604/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1604/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1605/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1605/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1606/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1606/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1607/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1607/33666 [01:00<20:01, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1608/33666 [01:00<20:01, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1608/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1609/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1609/33666 [01:00<20:01, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1610/33666 [01:00<20:01, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1610/33666 [01:00<20:01, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1611/33666 [01:00<20:01, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1611/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1612/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1612/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1613/33666 [01:00<20:01, 26.68it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▍         | 1613/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1614/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1614/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1615/33666 [01:00<20:01, 26.68it/s, loss=4.71, v_num=0]
Epoch 0:   5%|▍         | 1615/33666 [01:00<20:01, 26.68it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1616/33666 [01:00<20:01, 26.68it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1616/33666 [01:00<20:01, 26.68it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1617/33666 [01:00<20:01, 26.68it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▍         | 1617/33666 [01:00<20:01, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1618/33666 [01:00<20:00, 26.68it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▍         | 1618/33666 [01:00<20:00, 26.68it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▍         | 1619/33666 [01:00<20:00, 26.69it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▍         | 1619/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1620/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1620/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1621/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1621/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1622/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1622/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1623/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1623/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1624/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1624/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1625/33666 [01:00<20:00, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▍         | 1625/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1626/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1626/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1627/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1627/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1628/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1628/33666 [01:00<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1629/33666 [01:01<20:00, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1629/33666 [01:01<20:00, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1630/33666 [01:01<20:00, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1630/33666 [01:01<20:00, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1631/33666 [01:01<20:00, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1631/33666 [01:01<20:00, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1632/33666 [01:01<20:00, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1632/33666 [01:01<20:00, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1633/33666 [01:01<20:00, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1633/33666 [01:01<20:00, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1634/33666 [01:01<20:00, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1634/33666 [01:01<20:00, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1635/33666 [01:01<20:00, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1635/33666 [01:01<20:00, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1636/33666 [01:01<19:59, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1636/33666 [01:01<19:59, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1637/33666 [01:01<20:00, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1637/33666 [01:01<20:00, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1638/33666 [01:01<19:59, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1638/33666 [01:01<19:59, 26.69it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1639/33666 [01:01<19:59, 26.69it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1639/33666 [01:01<19:59, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1640/33666 [01:01<19:59, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1640/33666 [01:01<19:59, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1641/33666 [01:01<19:59, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1641/33666 [01:01<19:59, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1642/33666 [01:01<19:59, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1642/33666 [01:01<19:59, 26.69it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1643/33666 [01:01<19:59, 26.69it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1643/33666 [01:01<19:59, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1644/33666 [01:01<19:59, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1644/33666 [01:01<19:59, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1645/33666 [01:01<19:59, 26.69it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1645/33666 [01:01<19:59, 26.69it/s, loss=4.57, v_num=0]
Epoch 0:   5%|▍         | 1646/33666 [01:01<19:59, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   5%|▍         | 1646/33666 [01:01<19:59, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1647/33666 [01:01<19:59, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1647/33666 [01:01<19:59, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1648/33666 [01:01<19:59, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1648/33666 [01:01<19:59, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   5%|▍         | 1649/33666 [01:01<19:59, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   5%|▍         | 1649/33666 [01:01<19:59, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1650/33666 [01:01<19:59, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1650/33666 [01:01<19:59, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1651/33666 [01:01<19:59, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1651/33666 [01:01<19:59, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1652/33666 [01:01<19:58, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1652/33666 [01:01<19:58, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1653/33666 [01:01<19:58, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1653/33666 [01:01<19:58, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1654/33666 [01:01<19:58, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▍         | 1654/33666 [01:01<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1655/33666 [01:01<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1655/33666 [01:01<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1656/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1656/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1657/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1657/33666 [01:02<19:58, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1658/33666 [01:02<19:58, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1658/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1659/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1659/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1660/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1660/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1661/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1661/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1662/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1662/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1663/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1663/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1664/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1664/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1665/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1665/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1666/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1666/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1667/33666 [01:02<19:58, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▍         | 1667/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1668/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1668/33666 [01:02<19:58, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1669/33666 [01:02<19:58, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▍         | 1669/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1670/33666 [01:02<19:58, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▍         | 1670/33666 [01:02<19:58, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1671/33666 [01:02<19:58, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1671/33666 [01:02<19:58, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1672/33666 [01:02<19:58, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1672/33666 [01:02<19:58, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1673/33666 [01:02<19:58, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1673/33666 [01:02<19:58, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1674/33666 [01:02<19:58, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1674/33666 [01:02<19:58, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1675/33666 [01:02<19:58, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1675/33666 [01:02<19:58, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1676/33666 [01:02<19:58, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1676/33666 [01:02<19:58, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1677/33666 [01:02<19:58, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1677/33666 [01:02<19:58, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1678/33666 [01:02<19:58, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▍         | 1678/33666 [01:02<19:58, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1679/33666 [01:02<19:57, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1679/33666 [01:02<19:57, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1680/33666 [01:02<19:57, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▍         | 1680/33666 [01:02<19:57, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1681/33666 [01:02<19:57, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1681/33666 [01:02<19:57, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1682/33666 [01:02<19:57, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▍         | 1682/33666 [01:02<19:57, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1683/33666 [01:03<19:57, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▍         | 1683/33666 [01:03<19:57, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1684/33666 [01:03<19:57, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1684/33666 [01:03<19:57, 26.71it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1685/33666 [01:03<19:57, 26.71it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1685/33666 [01:03<19:57, 26.71it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1686/33666 [01:03<19:57, 26.71it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1686/33666 [01:03<19:57, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1687/33666 [01:03<19:57, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1687/33666 [01:03<19:57, 26.71it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1688/33666 [01:03<19:57, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1688/33666 [01:03<19:57, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1689/33666 [01:03<19:57, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1689/33666 [01:03<19:57, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1690/33666 [01:03<19:57, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1690/33666 [01:03<19:57, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1691/33666 [01:03<19:57, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1691/33666 [01:03<19:57, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1692/33666 [01:03<19:57, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1692/33666 [01:03<19:57, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1693/33666 [01:03<19:57, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1693/33666 [01:03<19:57, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1694/33666 [01:03<19:56, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1694/33666 [01:03<19:57, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1695/33666 [01:03<19:56, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1695/33666 [01:03<19:56, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1696/33666 [01:03<19:56, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1696/33666 [01:03<19:56, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1697/33666 [01:03<19:56, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1697/33666 [01:03<19:56, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1698/33666 [01:03<19:56, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1698/33666 [01:03<19:56, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1699/33666 [01:03<19:56, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1699/33666 [01:03<19:56, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▌         | 1700/33666 [01:03<19:56, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▌         | 1700/33666 [01:03<19:56, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1701/33666 [01:03<19:56, 26.72it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1701/33666 [01:03<19:56, 26.72it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▌         | 1702/33666 [01:03<19:56, 26.72it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▌         | 1702/33666 [01:03<19:56, 26.72it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▌         | 1703/33666 [01:03<19:56, 26.72it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▌         | 1703/33666 [01:03<19:56, 26.72it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▌         | 1704/33666 [01:03<19:56, 26.72it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▌         | 1704/33666 [01:03<19:56, 26.72it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▌         | 1705/33666 [01:03<19:56, 26.72it/s, loss=4.72, v_num=0]
Epoch 0:   5%|▌         | 1705/33666 [01:03<19:56, 26.72it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▌         | 1706/33666 [01:03<19:55, 26.73it/s, loss=4.7, v_num=0]
Epoch 0:   5%|▌         | 1706/33666 [01:03<19:55, 26.72it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1707/33666 [01:03<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1707/33666 [01:03<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1708/33666 [01:03<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1708/33666 [01:03<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1709/33666 [01:03<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1709/33666 [01:03<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1710/33666 [01:03<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1710/33666 [01:03<19:55, 26.73it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1711/33666 [01:04<19:55, 26.73it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1711/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1712/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1712/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1713/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1713/33666 [01:04<19:55, 26.73it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1714/33666 [01:04<19:55, 26.73it/s, loss=4.68, v_num=0]
Epoch 0:   5%|▌         | 1714/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1715/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1715/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1716/33666 [01:04<19:55, 26.73it/s, loss=4.69, v_num=0]
Epoch 0:   5%|▌         | 1716/33666 [01:04<19:55, 26.73it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1717/33666 [01:04<19:55, 26.73it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1717/33666 [01:04<19:55, 26.73it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1718/33666 [01:04<19:54, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1718/33666 [01:04<19:54, 26.74it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1719/33666 [01:04<19:54, 26.74it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1719/33666 [01:04<19:54, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1720/33666 [01:04<19:54, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1720/33666 [01:04<19:54, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1721/33666 [01:04<19:54, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   5%|▌         | 1721/33666 [01:04<19:54, 26.74it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1722/33666 [01:04<19:54, 26.74it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1722/33666 [01:04<19:54, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1723/33666 [01:04<19:54, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1723/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1724/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1724/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1725/33666 [01:04<19:54, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1725/33666 [01:04<19:54, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1726/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1726/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1727/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1727/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1728/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1728/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1729/33666 [01:04<19:54, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1729/33666 [01:04<19:54, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1730/33666 [01:04<19:54, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1730/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1731/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1731/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1732/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1732/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1733/33666 [01:04<19:54, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1733/33666 [01:04<19:54, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1734/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1734/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1735/33666 [01:04<19:54, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1735/33666 [01:04<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1736/33666 [01:04<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1736/33666 [01:04<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1737/33666 [01:04<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1737/33666 [01:04<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1738/33666 [01:05<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1738/33666 [01:05<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1739/33666 [01:05<19:54, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1739/33666 [01:05<19:54, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1740/33666 [01:05<19:54, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1740/33666 [01:05<19:54, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1741/33666 [01:05<19:54, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1741/33666 [01:05<19:54, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1742/33666 [01:05<19:54, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1742/33666 [01:05<19:54, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1743/33666 [01:05<19:54, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1743/33666 [01:05<19:54, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1744/33666 [01:05<19:54, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1744/33666 [01:05<19:54, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1745/33666 [01:05<19:54, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1745/33666 [01:05<19:54, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1746/33666 [01:05<19:53, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1746/33666 [01:05<19:53, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1747/33666 [01:05<19:54, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1747/33666 [01:05<19:54, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1748/33666 [01:05<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1748/33666 [01:05<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1749/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1749/33666 [01:05<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1750/33666 [01:05<19:53, 26.74it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1750/33666 [01:05<19:53, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1751/33666 [01:05<19:53, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1751/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1752/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1752/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1753/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1753/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1754/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1754/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1755/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1755/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1756/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1756/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1757/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1757/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1758/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1758/33666 [01:05<19:53, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1759/33666 [01:05<19:53, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1759/33666 [01:05<19:53, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1760/33666 [01:05<19:53, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1760/33666 [01:05<19:53, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1761/33666 [01:05<19:53, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1761/33666 [01:05<19:53, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1762/33666 [01:05<19:53, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1762/33666 [01:05<19:53, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1763/33666 [01:05<19:53, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1763/33666 [01:05<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1764/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1764/33666 [01:05<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1765/33666 [01:06<19:53, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1765/33666 [01:06<19:53, 26.74it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1766/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1766/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1767/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1767/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1768/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1768/33666 [01:06<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1769/33666 [01:06<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1769/33666 [01:06<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1770/33666 [01:06<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1770/33666 [01:06<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1771/33666 [01:06<19:53, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1771/33666 [01:06<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1772/33666 [01:06<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1772/33666 [01:06<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1773/33666 [01:06<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1773/33666 [01:06<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1774/33666 [01:06<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1774/33666 [01:06<19:53, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1775/33666 [01:06<19:53, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1775/33666 [01:06<19:53, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1776/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1776/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1777/33666 [01:06<19:53, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1777/33666 [01:06<19:53, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1778/33666 [01:06<19:53, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1778/33666 [01:06<19:53, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1779/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1779/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1780/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1780/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1781/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1781/33666 [01:06<19:53, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1782/33666 [01:06<19:52, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1782/33666 [01:06<19:52, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1783/33666 [01:06<19:52, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1783/33666 [01:06<19:52, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1784/33666 [01:06<19:52, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1784/33666 [01:06<19:52, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1785/33666 [01:06<19:52, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1785/33666 [01:06<19:52, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1786/33666 [01:06<19:53, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1786/33666 [01:06<19:53, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1787/33666 [01:06<19:52, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1787/33666 [01:06<19:53, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1788/33666 [01:06<19:52, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1788/33666 [01:06<19:52, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1789/33666 [01:06<19:52, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1789/33666 [01:06<19:52, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1790/33666 [01:06<19:52, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1790/33666 [01:06<19:52, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1791/33666 [01:07<19:52, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1791/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1792/33666 [01:07<19:53, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1792/33666 [01:07<19:53, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1793/33666 [01:07<19:53, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1793/33666 [01:07<19:53, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1794/33666 [01:07<19:53, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1794/33666 [01:07<19:53, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1795/33666 [01:07<19:53, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1795/33666 [01:07<19:53, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1796/33666 [01:07<19:52, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1796/33666 [01:07<19:52, 26.71it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1797/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1797/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1798/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1798/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1799/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1799/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1800/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1800/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1801/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1801/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1802/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1802/33666 [01:07<19:52, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1803/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1803/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1804/33666 [01:07<19:52, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1804/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1805/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1805/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1806/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1806/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1807/33666 [01:07<19:52, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   5%|▌         | 1807/33666 [01:07<19:52, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1808/33666 [01:07<19:52, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1808/33666 [01:07<19:52, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1809/33666 [01:07<19:52, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   5%|▌         | 1809/33666 [01:07<19:52, 26.72it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1810/33666 [01:07<19:52, 26.72it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1810/33666 [01:07<19:52, 26.72it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1811/33666 [01:07<19:51, 26.72it/s, loss=4.65, v_num=0]
Epoch 0:   5%|▌         | 1811/33666 [01:07<19:51, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1812/33666 [01:07<19:51, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1812/33666 [01:07<19:51, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1813/33666 [01:07<19:51, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1813/33666 [01:07<19:51, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1814/33666 [01:07<19:51, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1814/33666 [01:07<19:51, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1815/33666 [01:07<19:51, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1815/33666 [01:07<19:51, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1816/33666 [01:07<19:51, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1816/33666 [01:07<19:51, 26.74it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1817/33666 [01:07<19:51, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1817/33666 [01:07<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1818/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1818/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1819/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1819/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1820/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1820/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1821/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1821/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1822/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1822/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1823/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1823/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1824/33666 [01:08<19:51, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1824/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1825/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1825/33666 [01:08<19:51, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▌         | 1826/33666 [01:08<19:51, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▌         | 1826/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1827/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1827/33666 [01:08<19:51, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▌         | 1828/33666 [01:08<19:51, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▌         | 1828/33666 [01:08<19:51, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▌         | 1829/33666 [01:08<19:51, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   5%|▌         | 1829/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1830/33666 [01:08<19:51, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1830/33666 [01:08<19:51, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1831/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1831/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1832/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1832/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1833/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1833/33666 [01:08<19:51, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1834/33666 [01:08<19:51, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1834/33666 [01:08<19:51, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1835/33666 [01:08<19:51, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   5%|▌         | 1835/33666 [01:08<19:51, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1836/33666 [01:08<19:51, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   5%|▌         | 1836/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1837/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1837/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1838/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1838/33666 [01:08<19:51, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1839/33666 [01:08<19:51, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1839/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1840/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1840/33666 [01:08<19:51, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1841/33666 [01:08<19:51, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1841/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1842/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1842/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1843/33666 [01:08<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1843/33666 [01:08<19:51, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1844/33666 [01:09<19:51, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1844/33666 [01:09<19:51, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1845/33666 [01:09<19:51, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   5%|▌         | 1845/33666 [01:09<19:51, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1846/33666 [01:09<19:51, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1846/33666 [01:09<19:51, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1847/33666 [01:09<19:51, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1847/33666 [01:09<19:51, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1848/33666 [01:09<19:51, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1848/33666 [01:09<19:51, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1849/33666 [01:09<19:51, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1849/33666 [01:09<19:51, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1850/33666 [01:09<19:51, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   5%|▌         | 1850/33666 [01:09<19:51, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1851/33666 [01:09<19:51, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   5%|▌         | 1851/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1852/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1852/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1853/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1853/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1854/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1854/33666 [01:09<19:51, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1855/33666 [01:09<19:51, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1855/33666 [01:09<19:51, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1856/33666 [01:09<19:51, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1856/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1857/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1857/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1858/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1858/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1859/33666 [01:09<19:51, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1859/33666 [01:09<19:51, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1860/33666 [01:09<19:51, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1860/33666 [01:09<19:51, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1861/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1861/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1862/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1862/33666 [01:09<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1863/33666 [01:09<19:51, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1863/33666 [01:09<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1864/33666 [01:09<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1864/33666 [01:09<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1865/33666 [01:09<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1865/33666 [01:09<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1866/33666 [01:09<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1866/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1867/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1867/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1868/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1868/33666 [01:09<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1869/33666 [01:10<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1869/33666 [01:10<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1870/33666 [01:10<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1870/33666 [01:10<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1871/33666 [01:10<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1871/33666 [01:10<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1872/33666 [01:10<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1872/33666 [01:10<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1873/33666 [01:10<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1873/33666 [01:10<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1874/33666 [01:10<19:51, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1874/33666 [01:10<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1875/33666 [01:10<19:51, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1875/33666 [01:10<19:51, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1876/33666 [01:10<19:51, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1876/33666 [01:10<19:51, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1877/33666 [01:10<19:50, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1877/33666 [01:10<19:50, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1878/33666 [01:10<19:50, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1878/33666 [01:10<19:51, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1879/33666 [01:10<19:50, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1879/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1880/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1880/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1881/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1881/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1882/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1882/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1883/33666 [01:10<19:50, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1883/33666 [01:10<19:50, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1884/33666 [01:10<19:50, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1884/33666 [01:10<19:50, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1885/33666 [01:10<19:51, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1885/33666 [01:10<19:51, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1886/33666 [01:10<19:51, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1886/33666 [01:10<19:51, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1887/33666 [01:10<19:51, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1887/33666 [01:10<19:51, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1888/33666 [01:10<19:51, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1888/33666 [01:10<19:51, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1889/33666 [01:10<19:51, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1889/33666 [01:10<19:51, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1890/33666 [01:10<19:51, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1890/33666 [01:10<19:51, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1891/33666 [01:10<19:50, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1891/33666 [01:10<19:50, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1892/33666 [01:10<19:50, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1892/33666 [01:10<19:50, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▌         | 1893/33666 [01:10<19:50, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▌         | 1893/33666 [01:10<19:50, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▌         | 1894/33666 [01:10<19:50, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▌         | 1894/33666 [01:10<19:50, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1895/33666 [01:11<19:50, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1895/33666 [01:11<19:50, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1896/33666 [01:11<19:50, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1896/33666 [01:11<19:50, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1897/33666 [01:11<19:50, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1897/33666 [01:11<19:50, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1898/33666 [01:11<19:50, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1898/33666 [01:11<19:50, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1899/33666 [01:11<19:50, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1899/33666 [01:11<19:50, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1900/33666 [01:11<19:50, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1900/33666 [01:11<19:50, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1901/33666 [01:11<19:50, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1901/33666 [01:11<19:50, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1902/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1902/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1903/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1903/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1904/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1904/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1905/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1905/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1906/33666 [01:11<19:50, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1906/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1907/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1907/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1908/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1908/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1909/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1909/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1910/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1910/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1911/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1911/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1912/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1912/33666 [01:11<19:50, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   6%|▌         | 1913/33666 [01:11<19:50, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   6%|▌         | 1913/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1914/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1914/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1915/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1915/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1916/33666 [01:11<19:50, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1916/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1917/33666 [01:11<19:50, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1917/33666 [01:11<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1918/33666 [01:11<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1918/33666 [01:11<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1919/33666 [01:11<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1919/33666 [01:11<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1920/33666 [01:11<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1920/33666 [01:11<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1921/33666 [01:12<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1921/33666 [01:12<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1922/33666 [01:12<19:50, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1922/33666 [01:12<19:50, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1923/33666 [01:12<19:50, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1923/33666 [01:12<19:50, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1924/33666 [01:12<19:49, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1924/33666 [01:12<19:49, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1925/33666 [01:12<19:49, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1925/33666 [01:12<19:49, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1926/33666 [01:12<19:49, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1926/33666 [01:12<19:49, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1927/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1927/33666 [01:12<19:49, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1928/33666 [01:12<19:49, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▌         | 1928/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1929/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1929/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1930/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1930/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1931/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1931/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1932/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1932/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1933/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1933/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1934/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1934/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1935/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1935/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1936/33666 [01:12<19:49, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1936/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1937/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1937/33666 [01:12<19:49, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1938/33666 [01:12<19:49, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1938/33666 [01:12<19:49, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1939/33666 [01:12<19:49, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1939/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1940/33666 [01:12<19:49, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1940/33666 [01:12<19:49, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1941/33666 [01:12<19:48, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1941/33666 [01:12<19:48, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1942/33666 [01:12<19:48, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1942/33666 [01:12<19:48, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1943/33666 [01:12<19:49, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1943/33666 [01:12<19:49, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1944/33666 [01:12<19:48, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1944/33666 [01:12<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1945/33666 [01:12<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1945/33666 [01:12<19:48, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1946/33666 [01:12<19:48, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1946/33666 [01:12<19:48, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1947/33666 [01:12<19:48, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1947/33666 [01:12<19:48, 26.68it/s, loss=4.67, v_num=0]
Epoch 0:   6%|▌         | 1948/33666 [01:13<19:48, 26.68it/s, loss=4.67, v_num=0]
Epoch 0:   6%|▌         | 1948/33666 [01:13<19:48, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1949/33666 [01:13<19:48, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1949/33666 [01:13<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1950/33666 [01:13<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1950/33666 [01:13<19:48, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1951/33666 [01:13<19:48, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1951/33666 [01:13<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1952/33666 [01:13<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1952/33666 [01:13<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1953/33666 [01:13<19:48, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1953/33666 [01:13<19:48, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1954/33666 [01:13<19:48, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1954/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1955/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1955/33666 [01:13<19:48, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1956/33666 [01:13<19:48, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1956/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1957/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1957/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1958/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1958/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1959/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1959/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1960/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1960/33666 [01:13<19:48, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1961/33666 [01:13<19:48, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1961/33666 [01:13<19:48, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1962/33666 [01:13<19:48, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1962/33666 [01:13<19:48, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1963/33666 [01:13<19:48, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1963/33666 [01:13<19:48, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1964/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1964/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1965/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1965/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1966/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1966/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1967/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1967/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1968/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1968/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1969/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1969/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1970/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1970/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1971/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1971/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1972/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1972/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1973/33666 [01:13<19:47, 26.69it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 1973/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1974/33666 [01:13<19:47, 26.69it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 1974/33666 [01:13<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1975/33666 [01:13<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1975/33666 [01:13<19:47, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1976/33666 [01:14<19:47, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1976/33666 [01:14<19:47, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1977/33666 [01:14<19:47, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1977/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1978/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1978/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1979/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1979/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1980/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1980/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1981/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1981/33666 [01:14<19:47, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1982/33666 [01:14<19:47, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1982/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1983/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1983/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1984/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1984/33666 [01:14<19:47, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1985/33666 [01:14<19:47, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 1985/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1986/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1986/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1987/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1987/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1988/33666 [01:14<19:47, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 1988/33666 [01:14<19:47, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1989/33666 [01:14<19:47, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1989/33666 [01:14<19:47, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1990/33666 [01:14<19:47, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 1990/33666 [01:14<19:47, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1991/33666 [01:14<19:47, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1991/33666 [01:14<19:47, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1992/33666 [01:14<19:47, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1992/33666 [01:14<19:47, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1993/33666 [01:14<19:47, 26.68it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 1993/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1994/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1994/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1995/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1995/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1996/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1996/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1997/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1997/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1998/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1998/33666 [01:14<19:47, 26.68it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1999/33666 [01:14<19:47, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 1999/33666 [01:14<19:47, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2000/33666 [01:14<19:47, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2000/33666 [01:14<19:47, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2001/33666 [01:15<19:47, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2001/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2002/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2002/33666 [01:15<19:47, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2003/33666 [01:15<19:47, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2003/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2004/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2004/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2005/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2005/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2006/33666 [01:15<19:47, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2006/33666 [01:15<19:47, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2007/33666 [01:15<19:47, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2007/33666 [01:15<19:47, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2008/33666 [01:15<19:47, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2008/33666 [01:15<19:47, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2009/33666 [01:15<19:47, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2009/33666 [01:15<19:47, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2010/33666 [01:15<19:47, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   6%|▌         | 2010/33666 [01:15<19:47, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2011/33666 [01:15<19:47, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2011/33666 [01:15<19:47, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2012/33666 [01:15<19:47, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2012/33666 [01:15<19:47, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2013/33666 [01:15<19:47, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2013/33666 [01:15<19:47, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2014/33666 [01:15<19:47, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2014/33666 [01:15<19:47, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2015/33666 [01:15<19:47, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2015/33666 [01:15<19:47, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2016/33666 [01:15<19:47, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2016/33666 [01:15<19:47, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2017/33666 [01:15<19:47, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2017/33666 [01:15<19:47, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2018/33666 [01:15<19:47, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2018/33666 [01:15<19:47, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2019/33666 [01:15<19:47, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2019/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2020/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2020/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2021/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2021/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2022/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2022/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2023/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2023/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2024/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2024/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2025/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2025/33666 [01:15<19:47, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2026/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2026/33666 [01:16<19:47, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2027/33666 [01:16<19:47, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2027/33666 [01:16<19:47, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2028/33666 [01:16<19:47, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2028/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2029/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2029/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2030/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2030/33666 [01:16<19:47, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2031/33666 [01:16<19:47, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2031/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2032/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2032/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2033/33666 [01:16<19:47, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2033/33666 [01:16<19:47, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2034/33666 [01:16<19:47, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2034/33666 [01:16<19:47, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2035/33666 [01:16<19:47, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2035/33666 [01:16<19:47, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2036/33666 [01:16<19:47, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2036/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2037/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2037/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2038/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2038/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2039/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2039/33666 [01:16<19:47, 26.63it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2040/33666 [01:16<19:47, 26.63it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2040/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2041/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2041/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2042/33666 [01:16<19:47, 26.63it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2042/33666 [01:16<19:47, 26.63it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2043/33666 [01:16<19:47, 26.63it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2043/33666 [01:16<19:47, 26.63it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2044/33666 [01:16<19:47, 26.63it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2044/33666 [01:16<19:47, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2045/33666 [01:16<19:47, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2045/33666 [01:16<19:47, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2046/33666 [01:16<19:47, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2046/33666 [01:16<19:47, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2047/33666 [01:16<19:47, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2047/33666 [01:16<19:47, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2048/33666 [01:16<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2048/33666 [01:16<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2049/33666 [01:16<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2049/33666 [01:16<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2050/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2050/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2051/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2051/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2052/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2052/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2053/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2053/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2054/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2054/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2055/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2055/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2056/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2056/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2057/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2057/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2058/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2058/33666 [01:17<19:47, 26.62it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2059/33666 [01:17<19:47, 26.62it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2059/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2060/33666 [01:17<19:47, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2060/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2061/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2061/33666 [01:17<19:47, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2062/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2062/33666 [01:17<19:47, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2063/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2063/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2064/33666 [01:17<19:46, 26.62it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2064/33666 [01:17<19:46, 26.62it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2065/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2065/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2066/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2066/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2067/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2067/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2068/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2068/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2069/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2069/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2070/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2070/33666 [01:17<19:46, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2071/33666 [01:17<19:46, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2071/33666 [01:17<19:46, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2072/33666 [01:17<19:46, 26.63it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2072/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2073/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2073/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2074/33666 [01:17<19:46, 26.63it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2074/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2075/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2075/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2076/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2076/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2077/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2077/33666 [01:17<19:46, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2078/33666 [01:18<19:45, 26.63it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2078/33666 [01:18<19:45, 26.63it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 2079/33666 [01:18<19:45, 26.64it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▌         | 2079/33666 [01:18<19:45, 26.64it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2080/33666 [01:18<19:45, 26.64it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2080/33666 [01:18<19:45, 26.64it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2081/33666 [01:18<19:45, 26.64it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▌         | 2081/33666 [01:18<19:45, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2082/33666 [01:18<19:45, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2082/33666 [01:18<19:45, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2083/33666 [01:18<19:45, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2083/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2084/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2084/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2085/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2085/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2086/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2086/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2087/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2087/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2088/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2088/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2089/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2089/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2090/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2090/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2091/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2091/33666 [01:18<19:45, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2092/33666 [01:18<19:45, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▌         | 2092/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2093/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2093/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2094/33666 [01:18<19:45, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2094/33666 [01:18<19:45, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2095/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2095/33666 [01:18<19:44, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2096/33666 [01:18<19:44, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2096/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2097/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2097/33666 [01:18<19:44, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2098/33666 [01:18<19:44, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2098/33666 [01:18<19:44, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2099/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2099/33666 [01:18<19:44, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2100/33666 [01:18<19:44, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   6%|▌         | 2100/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2101/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2101/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2102/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2102/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2103/33666 [01:18<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▌         | 2103/33666 [01:18<19:44, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2104/33666 [01:18<19:44, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▌         | 2104/33666 [01:18<19:44, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2105/33666 [01:19<19:44, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2105/33666 [01:19<19:44, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2106/33666 [01:19<19:44, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2106/33666 [01:19<19:44, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2107/33666 [01:19<19:44, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2107/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2108/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2108/33666 [01:19<19:44, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2109/33666 [01:19<19:44, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2109/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2110/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2110/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2111/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2111/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2112/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2112/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2113/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2113/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2114/33666 [01:19<19:44, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2114/33666 [01:19<19:44, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2115/33666 [01:19<19:43, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2115/33666 [01:19<19:43, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2116/33666 [01:19<19:43, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2116/33666 [01:19<19:43, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2117/33666 [01:19<19:43, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2117/33666 [01:19<19:43, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2118/33666 [01:19<19:43, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2118/33666 [01:19<19:43, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2119/33666 [01:19<19:43, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2119/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2120/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2120/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2121/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2121/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2122/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2122/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2123/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2123/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2124/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2124/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2125/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2125/33666 [01:19<19:43, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▋         | 2126/33666 [01:19<19:43, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▋         | 2126/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2127/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2127/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2128/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2128/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2129/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2129/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2130/33666 [01:19<19:43, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2130/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2131/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2131/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2132/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2132/33666 [01:19<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2133/33666 [01:20<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2133/33666 [01:20<19:43, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2134/33666 [01:20<19:42, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2134/33666 [01:20<19:42, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2135/33666 [01:20<19:42, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2135/33666 [01:20<19:42, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▋         | 2136/33666 [01:20<19:42, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▋         | 2136/33666 [01:20<19:42, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▋         | 2137/33666 [01:20<19:42, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   6%|▋         | 2137/33666 [01:20<19:42, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2138/33666 [01:20<19:42, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2138/33666 [01:20<19:42, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2139/33666 [01:20<19:42, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2139/33666 [01:20<19:42, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2140/33666 [01:20<19:42, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2140/33666 [01:20<19:42, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2141/33666 [01:20<19:42, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2141/33666 [01:20<19:42, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2142/33666 [01:20<19:41, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2142/33666 [01:20<19:41, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2143/33666 [01:20<19:41, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   6%|▋         | 2143/33666 [01:20<19:41, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▋         | 2144/33666 [01:20<19:41, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▋         | 2144/33666 [01:20<19:41, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▋         | 2145/33666 [01:20<19:41, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   6%|▋         | 2145/33666 [01:20<19:41, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2146/33666 [01:20<19:41, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2146/33666 [01:20<19:41, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2147/33666 [01:20<19:41, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2147/33666 [01:20<19:41, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2148/33666 [01:20<19:41, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2148/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2149/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2149/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2150/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2150/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2151/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2151/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2152/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2152/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2153/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2153/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2154/33666 [01:20<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2154/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2155/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2155/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2156/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2156/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2157/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2157/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2158/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2158/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2159/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2159/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2160/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2160/33666 [01:20<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2161/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2161/33666 [01:21<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2162/33666 [01:21<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2162/33666 [01:21<19:41, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2163/33666 [01:21<19:41, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   6%|▋         | 2163/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2164/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2164/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2165/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2165/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2166/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2166/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2167/33666 [01:21<19:41, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2167/33666 [01:21<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2168/33666 [01:21<19:41, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2168/33666 [01:21<19:41, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2169/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2169/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2170/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2170/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2171/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2171/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2172/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2172/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2173/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2173/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2174/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2174/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2175/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2175/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2176/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2176/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2177/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2177/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2178/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2178/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2179/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2179/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2180/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2180/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2181/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2181/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2182/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2182/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2183/33666 [01:21<19:40, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   6%|▋         | 2183/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2184/33666 [01:21<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2184/33666 [01:21<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2185/33666 [01:21<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2185/33666 [01:21<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2186/33666 [01:21<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2186/33666 [01:21<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2187/33666 [01:22<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   6%|▋         | 2187/33666 [01:22<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2188/33666 [01:22<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   6%|▋         | 2188/33666 [01:22<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2189/33666 [01:22<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2189/33666 [01:22<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2190/33666 [01:22<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2190/33666 [01:22<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2191/33666 [01:22<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2191/33666 [01:22<19:40, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2192/33666 [01:22<19:40, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2192/33666 [01:22<19:40, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2193/33666 [01:22<19:40, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2193/33666 [01:22<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2194/33666 [01:22<19:40, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2194/33666 [01:22<19:40, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2195/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2195/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2196/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2196/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2197/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2197/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2198/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2198/33666 [01:22<19:39, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2199/33666 [01:22<19:39, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2199/33666 [01:22<19:39, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2200/33666 [01:22<19:39, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2200/33666 [01:22<19:39, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2201/33666 [01:22<19:39, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2201/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2202/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2202/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2203/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2203/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2204/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2204/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2205/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2205/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2206/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2206/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2207/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2207/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2208/33666 [01:22<19:39, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2208/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2209/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2209/33666 [01:22<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2210/33666 [01:22<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2210/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2211/33666 [01:22<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2211/33666 [01:22<19:39, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2212/33666 [01:22<19:39, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2212/33666 [01:22<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2213/33666 [01:22<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2213/33666 [01:22<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2214/33666 [01:23<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2214/33666 [01:23<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2215/33666 [01:23<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2215/33666 [01:23<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2216/33666 [01:23<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2216/33666 [01:23<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2217/33666 [01:23<19:39, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2217/33666 [01:23<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2218/33666 [01:23<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2218/33666 [01:23<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2219/33666 [01:23<19:39, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2219/33666 [01:23<19:39, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2220/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2220/33666 [01:23<19:39, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2221/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2221/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2222/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2222/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2223/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2223/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2224/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2224/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2225/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2225/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2226/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2226/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2227/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2227/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2228/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2228/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2229/33666 [01:23<19:38, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2229/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2230/33666 [01:23<19:38, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2230/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2231/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2231/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2232/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2232/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2233/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2233/33666 [01:23<19:38, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2234/33666 [01:23<19:38, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2234/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2235/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2235/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2236/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2236/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2237/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2237/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2238/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2238/33666 [01:23<19:38, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2239/33666 [01:23<19:38, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2239/33666 [01:23<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2240/33666 [01:24<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2240/33666 [01:24<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2241/33666 [01:24<19:38, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2241/33666 [01:24<19:38, 26.66it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2242/33666 [01:24<19:38, 26.66it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2242/33666 [01:24<19:38, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2243/33666 [01:24<19:38, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2243/33666 [01:24<19:38, 26.66it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2244/33666 [01:24<19:38, 26.66it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2244/33666 [01:24<19:38, 26.66it/s, loss=4.53, v_num=0]
Epoch 0:   7%|▋         | 2245/33666 [01:24<19:38, 26.67it/s, loss=4.53, v_num=0]
Epoch 0:   7%|▋         | 2245/33666 [01:24<19:38, 26.67it/s, loss=4.52, v_num=0]
Epoch 0:   7%|▋         | 2246/33666 [01:24<19:38, 26.67it/s, loss=4.52, v_num=0]
Epoch 0:   7%|▋         | 2246/33666 [01:24<19:38, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2247/33666 [01:24<19:38, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2247/33666 [01:24<19:38, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2248/33666 [01:24<19:38, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2248/33666 [01:24<19:38, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2249/33666 [01:24<19:37, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2249/33666 [01:24<19:37, 26.67it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2250/33666 [01:24<19:37, 26.67it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2250/33666 [01:24<19:37, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2251/33666 [01:24<19:37, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2251/33666 [01:24<19:37, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2252/33666 [01:24<19:37, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2252/33666 [01:24<19:37, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2253/33666 [01:24<19:37, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2253/33666 [01:24<19:37, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2254/33666 [01:24<19:37, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2254/33666 [01:24<19:37, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2255/33666 [01:24<19:37, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2255/33666 [01:24<19:37, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2256/33666 [01:24<19:37, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2256/33666 [01:24<19:37, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2257/33666 [01:24<19:37, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2257/33666 [01:24<19:37, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2258/33666 [01:24<19:37, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2258/33666 [01:24<19:37, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2259/33666 [01:24<19:37, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2259/33666 [01:24<19:37, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2260/33666 [01:24<19:37, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2260/33666 [01:24<19:37, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2261/33666 [01:24<19:37, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2261/33666 [01:24<19:37, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2262/33666 [01:24<19:37, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2262/33666 [01:24<19:37, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2263/33666 [01:24<19:37, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2263/33666 [01:24<19:37, 26.67it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2264/33666 [01:24<19:37, 26.67it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2264/33666 [01:24<19:37, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2265/33666 [01:24<19:37, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2265/33666 [01:24<19:37, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   7%|▋         | 2266/33666 [01:24<19:37, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   7%|▋         | 2266/33666 [01:24<19:37, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2267/33666 [01:25<19:37, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2267/33666 [01:25<19:37, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2268/33666 [01:25<19:37, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2268/33666 [01:25<19:37, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   7%|▋         | 2269/33666 [01:25<19:37, 26.66it/s, loss=4.71, v_num=0]
Epoch 0:   7%|▋         | 2269/33666 [01:25<19:37, 26.66it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2270/33666 [01:25<19:37, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2270/33666 [01:25<19:37, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2271/33666 [01:25<19:37, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2271/33666 [01:25<19:37, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2272/33666 [01:25<19:37, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2272/33666 [01:25<19:37, 26.67it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2273/33666 [01:25<19:37, 26.67it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2273/33666 [01:25<19:37, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2274/33666 [01:25<19:37, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2274/33666 [01:25<19:37, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2275/33666 [01:25<19:37, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2275/33666 [01:25<19:37, 26.66it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2276/33666 [01:25<19:37, 26.66it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2276/33666 [01:25<19:37, 26.66it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2277/33666 [01:25<19:37, 26.66it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2277/33666 [01:25<19:37, 26.66it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2278/33666 [01:25<19:37, 26.66it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2278/33666 [01:25<19:37, 26.66it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2279/33666 [01:25<19:37, 26.66it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2279/33666 [01:25<19:37, 26.66it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2280/33666 [01:25<19:37, 26.66it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2280/33666 [01:25<19:37, 26.66it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2281/33666 [01:25<19:37, 26.66it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2281/33666 [01:25<19:37, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2282/33666 [01:25<19:37, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2282/33666 [01:25<19:37, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2283/33666 [01:25<19:37, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2283/33666 [01:25<19:37, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2284/33666 [01:25<19:37, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2284/33666 [01:25<19:37, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2285/33666 [01:25<19:36, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2285/33666 [01:25<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2286/33666 [01:25<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2286/33666 [01:25<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2287/33666 [01:25<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2287/33666 [01:25<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2288/33666 [01:25<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2288/33666 [01:25<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2289/33666 [01:25<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2289/33666 [01:25<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2290/33666 [01:25<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2290/33666 [01:25<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2291/33666 [01:25<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2291/33666 [01:25<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2292/33666 [01:25<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2292/33666 [01:25<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2293/33666 [01:26<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2293/33666 [01:26<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2294/33666 [01:26<19:36, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2294/33666 [01:26<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2295/33666 [01:26<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2295/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2296/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2296/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2297/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2297/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2298/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2298/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2299/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2299/33666 [01:26<19:36, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2300/33666 [01:26<19:36, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2300/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2301/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2301/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2302/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2302/33666 [01:26<19:36, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2303/33666 [01:26<19:36, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2303/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2304/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2304/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2305/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2305/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2306/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2306/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2307/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2307/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2308/33666 [01:26<19:36, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2308/33666 [01:26<19:36, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2309/33666 [01:26<19:36, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2309/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2310/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2310/33666 [01:26<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2311/33666 [01:26<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2311/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2312/33666 [01:26<19:36, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2312/33666 [01:26<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2313/33666 [01:26<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2313/33666 [01:26<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2314/33666 [01:26<19:36, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2314/33666 [01:26<19:36, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2315/33666 [01:26<19:35, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2315/33666 [01:26<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2316/33666 [01:26<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2316/33666 [01:26<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2317/33666 [01:26<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2317/33666 [01:26<19:35, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2318/33666 [01:26<19:35, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2318/33666 [01:26<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2319/33666 [01:26<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2319/33666 [01:26<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2320/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2320/33666 [01:27<19:35, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2321/33666 [01:27<19:35, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2321/33666 [01:27<19:35, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2322/33666 [01:27<19:35, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2322/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2323/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2323/33666 [01:27<19:35, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2324/33666 [01:27<19:35, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2324/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2325/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2325/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2326/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2326/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2327/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2327/33666 [01:27<19:35, 26.66it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2328/33666 [01:27<19:35, 26.66it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2328/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2329/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2329/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2330/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2330/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2331/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2331/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2332/33666 [01:27<19:35, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2332/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2333/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2333/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2334/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2334/33666 [01:27<19:35, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2335/33666 [01:27<19:35, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2335/33666 [01:27<19:35, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2336/33666 [01:27<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2336/33666 [01:27<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2337/33666 [01:27<19:35, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2337/33666 [01:27<19:35, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2338/33666 [01:27<19:34, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2338/33666 [01:27<19:34, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2339/33666 [01:27<19:34, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2339/33666 [01:27<19:34, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2340/33666 [01:27<19:34, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2340/33666 [01:27<19:34, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2341/33666 [01:27<19:34, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2341/33666 [01:27<19:34, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2342/33666 [01:27<19:34, 26.66it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2342/33666 [01:27<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2343/33666 [01:27<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2343/33666 [01:27<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2344/33666 [01:27<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2344/33666 [01:27<19:34, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2345/33666 [01:27<19:34, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2345/33666 [01:27<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2346/33666 [01:27<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2346/33666 [01:27<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2347/33666 [01:28<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2347/33666 [01:28<19:34, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2348/33666 [01:28<19:34, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2348/33666 [01:28<19:34, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2349/33666 [01:28<19:34, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2349/33666 [01:28<19:34, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2350/33666 [01:28<19:34, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2350/33666 [01:28<19:34, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2351/33666 [01:28<19:34, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2351/33666 [01:28<19:34, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2352/33666 [01:28<19:34, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2352/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2353/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2353/33666 [01:28<19:34, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2354/33666 [01:28<19:34, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2354/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2355/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2355/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2356/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2356/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2357/33666 [01:28<19:34, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2357/33666 [01:28<19:34, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2358/33666 [01:28<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2358/33666 [01:28<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2359/33666 [01:28<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2359/33666 [01:28<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2360/33666 [01:28<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2360/33666 [01:28<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2361/33666 [01:28<19:34, 26.66it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2361/33666 [01:28<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2362/33666 [01:28<19:34, 26.66it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2362/33666 [01:28<19:34, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2363/33666 [01:28<19:34, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2363/33666 [01:28<19:34, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2364/33666 [01:28<19:34, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2364/33666 [01:28<19:34, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2365/33666 [01:28<19:34, 26.66it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2365/33666 [01:28<19:34, 26.66it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2366/33666 [01:28<19:34, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2366/33666 [01:28<19:34, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2367/33666 [01:28<19:34, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2367/33666 [01:28<19:34, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2368/33666 [01:28<19:34, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2368/33666 [01:28<19:34, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2369/33666 [01:28<19:34, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2369/33666 [01:28<19:34, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2370/33666 [01:28<19:34, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2370/33666 [01:28<19:34, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2371/33666 [01:28<19:34, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2371/33666 [01:28<19:34, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2372/33666 [01:29<19:34, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2372/33666 [01:29<19:34, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2373/33666 [01:29<19:34, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2373/33666 [01:29<19:34, 26.65it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2374/33666 [01:29<19:34, 26.64it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2374/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2375/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2375/33666 [01:29<19:34, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2376/33666 [01:29<19:34, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2376/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2377/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2377/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2378/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2378/33666 [01:29<19:34, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2379/33666 [01:29<19:34, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2379/33666 [01:29<19:34, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2380/33666 [01:29<19:34, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2380/33666 [01:29<19:34, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2381/33666 [01:29<19:34, 26.64it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2381/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2382/33666 [01:29<19:34, 26.64it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2382/33666 [01:29<19:34, 26.64it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2383/33666 [01:29<19:34, 26.64it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2383/33666 [01:29<19:34, 26.64it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2384/33666 [01:29<19:34, 26.64it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2384/33666 [01:29<19:34, 26.64it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2385/33666 [01:29<19:34, 26.64it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2385/33666 [01:29<19:34, 26.64it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2386/33666 [01:29<19:34, 26.64it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2386/33666 [01:29<19:34, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2387/33666 [01:29<19:33, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2387/33666 [01:29<19:33, 26.64it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2388/33666 [01:29<19:33, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2388/33666 [01:29<19:33, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2389/33666 [01:29<19:33, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2389/33666 [01:29<19:33, 26.64it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2390/33666 [01:29<19:33, 26.65it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2390/33666 [01:29<19:33, 26.65it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2391/33666 [01:29<19:33, 26.65it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2391/33666 [01:29<19:33, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2392/33666 [01:29<19:33, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2392/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2393/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2393/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2394/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2394/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2395/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2395/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2396/33666 [01:29<19:33, 26.65it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2396/33666 [01:29<19:33, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2397/33666 [01:29<19:33, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2397/33666 [01:29<19:33, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2398/33666 [01:29<19:33, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2398/33666 [01:29<19:33, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2399/33666 [01:30<19:33, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2399/33666 [01:30<19:33, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2400/33666 [01:30<19:33, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2400/33666 [01:30<19:33, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2401/33666 [01:30<19:33, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2401/33666 [01:30<19:33, 26.65it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2402/33666 [01:30<19:33, 26.65it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2402/33666 [01:30<19:33, 26.65it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2403/33666 [01:30<19:32, 26.65it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2403/33666 [01:30<19:32, 26.65it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2404/33666 [01:30<19:32, 26.65it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2404/33666 [01:30<19:32, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2405/33666 [01:30<19:32, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2405/33666 [01:30<19:32, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2406/33666 [01:30<19:32, 26.65it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2406/33666 [01:30<19:32, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2407/33666 [01:30<19:32, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2407/33666 [01:30<19:32, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2408/33666 [01:30<19:32, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2408/33666 [01:30<19:32, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2409/33666 [01:30<19:32, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2409/33666 [01:30<19:32, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2410/33666 [01:30<19:32, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2410/33666 [01:30<19:32, 26.65it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2411/33666 [01:30<19:32, 26.65it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2411/33666 [01:30<19:32, 26.65it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2412/33666 [01:30<19:32, 26.65it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2412/33666 [01:30<19:32, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2413/33666 [01:30<19:32, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2413/33666 [01:30<19:32, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2414/33666 [01:30<19:32, 26.65it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2414/33666 [01:30<19:32, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2415/33666 [01:30<19:32, 26.65it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2415/33666 [01:30<19:32, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2416/33666 [01:30<19:32, 26.65it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2416/33666 [01:30<19:32, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2417/33666 [01:30<19:32, 26.65it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2417/33666 [01:30<19:32, 26.65it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2418/33666 [01:30<19:32, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2418/33666 [01:30<19:32, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2419/33666 [01:30<19:32, 26.66it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2419/33666 [01:30<19:32, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2420/33666 [01:30<19:32, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2420/33666 [01:30<19:32, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2421/33666 [01:30<19:32, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2421/33666 [01:30<19:32, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2422/33666 [01:30<19:32, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2422/33666 [01:30<19:32, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2423/33666 [01:30<19:32, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2423/33666 [01:30<19:32, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2424/33666 [01:30<19:32, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2424/33666 [01:30<19:32, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2425/33666 [01:30<19:31, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2425/33666 [01:30<19:31, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2426/33666 [01:31<19:31, 26.66it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2426/33666 [01:31<19:31, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2427/33666 [01:31<19:31, 26.66it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2427/33666 [01:31<19:31, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2428/33666 [01:31<19:31, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2428/33666 [01:31<19:31, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2429/33666 [01:31<19:31, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2429/33666 [01:31<19:31, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2430/33666 [01:31<19:31, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2430/33666 [01:31<19:31, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2431/33666 [01:31<19:31, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2431/33666 [01:31<19:31, 26.66it/s, loss=4.53, v_num=0]
Epoch 0:   7%|▋         | 2432/33666 [01:31<19:31, 26.66it/s, loss=4.53, v_num=0]
Epoch 0:   7%|▋         | 2432/33666 [01:31<19:31, 26.66it/s, loss=4.51, v_num=0]
Epoch 0:   7%|▋         | 2433/33666 [01:31<19:31, 26.66it/s, loss=4.51, v_num=0]
Epoch 0:   7%|▋         | 2433/33666 [01:31<19:31, 26.66it/s, loss=4.52, v_num=0]
Epoch 0:   7%|▋         | 2434/33666 [01:31<19:31, 26.66it/s, loss=4.52, v_num=0]
Epoch 0:   7%|▋         | 2434/33666 [01:31<19:31, 26.66it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2435/33666 [01:31<19:31, 26.66it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2435/33666 [01:31<19:31, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2436/33666 [01:31<19:31, 26.66it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2436/33666 [01:31<19:31, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2437/33666 [01:31<19:31, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2437/33666 [01:31<19:31, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2438/33666 [01:31<19:31, 26.66it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2438/33666 [01:31<19:31, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2439/33666 [01:31<19:31, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2439/33666 [01:31<19:31, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2440/33666 [01:31<19:31, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2440/33666 [01:31<19:31, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2441/33666 [01:31<19:31, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2441/33666 [01:31<19:31, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2442/33666 [01:31<19:30, 26.67it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2442/33666 [01:31<19:30, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2443/33666 [01:31<19:30, 26.67it/s, loss=4.54, v_num=0]
Epoch 0:   7%|▋         | 2443/33666 [01:31<19:30, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2444/33666 [01:31<19:30, 26.66it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2444/33666 [01:31<19:30, 26.66it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2445/33666 [01:31<19:30, 26.67it/s, loss=4.55, v_num=0]
Epoch 0:   7%|▋         | 2445/33666 [01:31<19:30, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2446/33666 [01:31<19:30, 26.67it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2446/33666 [01:31<19:30, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2447/33666 [01:31<19:30, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2447/33666 [01:31<19:30, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2448/33666 [01:31<19:30, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2448/33666 [01:31<19:30, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2449/33666 [01:31<19:30, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2449/33666 [01:31<19:30, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2450/33666 [01:31<19:30, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2450/33666 [01:31<19:30, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2451/33666 [01:31<19:30, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2451/33666 [01:31<19:30, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2452/33666 [01:31<19:30, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2452/33666 [01:31<19:30, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2453/33666 [01:31<19:30, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2453/33666 [01:31<19:30, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2454/33666 [01:32<19:30, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2454/33666 [01:32<19:30, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2455/33666 [01:32<19:30, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2455/33666 [01:32<19:30, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2456/33666 [01:32<19:30, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2456/33666 [01:32<19:30, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2457/33666 [01:32<19:30, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2457/33666 [01:32<19:30, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2458/33666 [01:32<19:30, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2458/33666 [01:32<19:30, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2459/33666 [01:32<19:30, 26.67it/s, loss=4.67, v_num=0]
Epoch 0:   7%|▋         | 2459/33666 [01:32<19:30, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2460/33666 [01:32<19:30, 26.67it/s, loss=4.69, v_num=0]
Epoch 0:   7%|▋         | 2460/33666 [01:32<19:30, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   7%|▋         | 2461/33666 [01:32<19:30, 26.67it/s, loss=4.71, v_num=0]
Epoch 0:   7%|▋         | 2461/33666 [01:32<19:30, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2462/33666 [01:32<19:30, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2462/33666 [01:32<19:30, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2463/33666 [01:32<19:30, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2463/33666 [01:32<19:30, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2464/33666 [01:32<19:29, 26.67it/s, loss=4.7, v_num=0]
Epoch 0:   7%|▋         | 2464/33666 [01:32<19:29, 26.67it/s, loss=4.72, v_num=0]
Epoch 0:   7%|▋         | 2465/33666 [01:32<19:29, 26.67it/s, loss=4.72, v_num=0]
Epoch 0:   7%|▋         | 2465/33666 [01:32<19:29, 26.67it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2466/33666 [01:32<19:29, 26.67it/s, loss=4.68, v_num=0]
Epoch 0:   7%|▋         | 2466/33666 [01:32<19:29, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2467/33666 [01:32<19:29, 26.67it/s, loss=4.66, v_num=0]
Epoch 0:   7%|▋         | 2467/33666 [01:32<19:29, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2468/33666 [01:32<19:29, 26.67it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2468/33666 [01:32<19:29, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2469/33666 [01:32<19:29, 26.67it/s, loss=4.65, v_num=0]
Epoch 0:   7%|▋         | 2469/33666 [01:32<19:29, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2470/33666 [01:32<19:29, 26.67it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2470/33666 [01:32<19:29, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2471/33666 [01:32<19:29, 26.67it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2471/33666 [01:32<19:29, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2472/33666 [01:32<19:29, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2472/33666 [01:32<19:29, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2473/33666 [01:32<19:29, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2473/33666 [01:32<19:29, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2474/33666 [01:32<19:29, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2474/33666 [01:32<19:29, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2475/33666 [01:32<19:29, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2475/33666 [01:32<19:29, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2476/33666 [01:32<19:29, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2476/33666 [01:32<19:29, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2477/33666 [01:32<19:29, 26.67it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2477/33666 [01:32<19:29, 26.67it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2478/33666 [01:32<19:29, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2478/33666 [01:32<19:29, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2479/33666 [01:32<19:29, 26.67it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2479/33666 [01:32<19:29, 26.67it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2480/33666 [01:32<19:29, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2480/33666 [01:32<19:29, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2481/33666 [01:33<19:29, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2481/33666 [01:33<19:29, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2482/33666 [01:33<19:28, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2482/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2483/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2483/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2484/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2484/33666 [01:33<19:28, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2485/33666 [01:33<19:28, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2485/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2486/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2486/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2487/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2487/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2488/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2488/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2489/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2489/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2490/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2490/33666 [01:33<19:28, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2491/33666 [01:33<19:28, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2491/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2492/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2492/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2493/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2493/33666 [01:33<19:28, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2494/33666 [01:33<19:28, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2494/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2495/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2495/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2496/33666 [01:33<19:28, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   7%|▋         | 2496/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2497/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2497/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2498/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2498/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2499/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2499/33666 [01:33<19:28, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2500/33666 [01:33<19:28, 26.68it/s, loss=4.62, v_num=0]
Epoch 0:   7%|▋         | 2500/33666 [01:33<19:28, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2501/33666 [01:33<19:28, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2501/33666 [01:33<19:28, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2502/33666 [01:33<19:28, 26.68it/s, loss=4.64, v_num=0]
Epoch 0:   7%|▋         | 2502/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2503/33666 [01:33<19:28, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2503/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2504/33666 [01:33<19:28, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2504/33666 [01:33<19:28, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2505/33666 [01:33<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2505/33666 [01:33<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2506/33666 [01:33<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2506/33666 [01:33<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2507/33666 [01:33<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2507/33666 [01:33<19:27, 26.68it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2508/33666 [01:33<19:27, 26.68it/s, loss=4.56, v_num=0]
Epoch 0:   7%|▋         | 2508/33666 [01:33<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2509/33666 [01:34<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2509/33666 [01:34<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2510/33666 [01:34<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2510/33666 [01:34<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2511/33666 [01:34<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2511/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2512/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2512/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2513/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2513/33666 [01:34<19:27, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2514/33666 [01:34<19:27, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2514/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2515/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2515/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2516/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2516/33666 [01:34<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2517/33666 [01:34<19:27, 26.68it/s, loss=4.57, v_num=0]
Epoch 0:   7%|▋         | 2517/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2518/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2518/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2519/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2519/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2520/33666 [01:34<19:27, 26.68it/s, loss=4.59, v_num=0]
Epoch 0:   7%|▋         | 2520/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2521/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2521/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2522/33666 [01:34<19:27, 26.68it/s, loss=4.58, v_num=0]
Epoch 0:   7%|▋         | 2522/33666 [01:34<19:27, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2523/33666 [01:34<19:27, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   7%|▋         | 2523/33666 [01:34<19:27, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2524/33666 [01:34<19:27, 26.68it/s, loss=4.61, v_num=0]
Epoch 0:   7%|▋         | 2524/33666 [01:34<19:27, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2525/33666 [01:34<19:26, 26.68it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2525/33666 [01:34<19:26, 26.68it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2526/33666 [01:34<19:26, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2526/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2527/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2527/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2528/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2528/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2529/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2529/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2530/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2530/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2531/33666 [01:34<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2531/33666 [01:34<19:26, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2532/33666 [01:34<19:26, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2532/33666 [01:34<19:26, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2533/33666 [01:34<19:26, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2533/33666 [01:34<19:26, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2534/33666 [01:34<19:26, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2534/33666 [01:34<19:26, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2535/33666 [01:34<19:26, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2535/33666 [01:34<19:26, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2536/33666 [01:35<19:26, 26.69it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2536/33666 [01:35<19:26, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2537/33666 [01:35<19:26, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2537/33666 [01:35<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2538/33666 [01:35<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2538/33666 [01:35<19:26, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2539/33666 [01:35<19:26, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2539/33666 [01:35<19:26, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2540/33666 [01:35<19:26, 26.69it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2540/33666 [01:35<19:26, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2541/33666 [01:35<19:26, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2541/33666 [01:35<19:26, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2542/33666 [01:35<19:25, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2542/33666 [01:35<19:25, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2543/33666 [01:35<19:25, 26.69it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2543/33666 [01:35<19:25, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2544/33666 [01:35<19:25, 26.69it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2544/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2545/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2545/33666 [01:35<19:25, 26.69it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2546/33666 [01:35<19:25, 26.69it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2546/33666 [01:35<19:25, 26.69it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2547/33666 [01:35<19:25, 26.70it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2547/33666 [01:35<19:25, 26.70it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2548/33666 [01:35<19:25, 26.69it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2548/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2549/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2549/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2550/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2550/33666 [01:35<19:25, 26.69it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2551/33666 [01:35<19:25, 26.70it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2551/33666 [01:35<19:25, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2552/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2552/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2553/33666 [01:35<19:25, 26.69it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2553/33666 [01:35<19:25, 26.69it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2554/33666 [01:35<19:25, 26.69it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2554/33666 [01:35<19:25, 26.69it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2555/33666 [01:35<19:25, 26.69it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2555/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2556/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2556/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2557/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2557/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2558/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2558/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2559/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2559/33666 [01:35<19:25, 26.69it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2560/33666 [01:35<19:25, 26.69it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2560/33666 [01:35<19:25, 26.69it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2561/33666 [01:35<19:25, 26.70it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2561/33666 [01:35<19:25, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2562/33666 [01:35<19:25, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2562/33666 [01:35<19:25, 26.70it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2563/33666 [01:36<19:25, 26.70it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2563/33666 [01:36<19:25, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2564/33666 [01:36<19:25, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2564/33666 [01:36<19:25, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2565/33666 [01:36<19:24, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2565/33666 [01:36<19:24, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2566/33666 [01:36<19:24, 26.70it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2566/33666 [01:36<19:24, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2567/33666 [01:36<19:24, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2567/33666 [01:36<19:24, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2568/33666 [01:36<19:24, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2568/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2569/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2569/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2570/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2570/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2571/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2571/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2572/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2572/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2573/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2573/33666 [01:36<19:24, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2574/33666 [01:36<19:24, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2574/33666 [01:36<19:24, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2575/33666 [01:36<19:24, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2575/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2576/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2576/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2577/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2577/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2578/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2578/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2579/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2579/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2580/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2580/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2581/33666 [01:36<19:24, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2581/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2582/33666 [01:36<19:24, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2582/33666 [01:36<19:24, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2583/33666 [01:36<19:24, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2583/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2584/33666 [01:36<19:24, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2584/33666 [01:36<19:24, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2585/33666 [01:36<19:23, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2585/33666 [01:36<19:23, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2586/33666 [01:36<19:23, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2586/33666 [01:36<19:23, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2587/33666 [01:36<19:23, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2587/33666 [01:36<19:23, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2588/33666 [01:36<19:23, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2588/33666 [01:36<19:23, 26.70it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2589/33666 [01:36<19:23, 26.70it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2589/33666 [01:36<19:23, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2590/33666 [01:36<19:23, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2590/33666 [01:36<19:23, 26.70it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2591/33666 [01:37<19:23, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2591/33666 [01:37<19:23, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2592/33666 [01:37<19:23, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2592/33666 [01:37<19:23, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2593/33666 [01:37<19:23, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2593/33666 [01:37<19:23, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2594/33666 [01:37<19:23, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2594/33666 [01:37<19:23, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2595/33666 [01:37<19:23, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2595/33666 [01:37<19:23, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2596/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2596/33666 [01:37<19:23, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2597/33666 [01:37<19:23, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2597/33666 [01:37<19:23, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2598/33666 [01:37<19:23, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2598/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2599/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2599/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2600/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2600/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2601/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2601/33666 [01:37<19:23, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2602/33666 [01:37<19:23, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2602/33666 [01:37<19:23, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2603/33666 [01:37<19:23, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2603/33666 [01:37<19:23, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2604/33666 [01:37<19:23, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2604/33666 [01:37<19:23, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2605/33666 [01:37<19:23, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2605/33666 [01:37<19:23, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2606/33666 [01:37<19:22, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2606/33666 [01:37<19:22, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2607/33666 [01:37<19:22, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2607/33666 [01:37<19:22, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2608/33666 [01:37<19:22, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2608/33666 [01:37<19:22, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2609/33666 [01:37<19:22, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2609/33666 [01:37<19:22, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2610/33666 [01:37<19:22, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2610/33666 [01:37<19:22, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2611/33666 [01:37<19:22, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2611/33666 [01:37<19:22, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2612/33666 [01:37<19:22, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2612/33666 [01:37<19:22, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2613/33666 [01:37<19:22, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2613/33666 [01:37<19:22, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2614/33666 [01:37<19:22, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2614/33666 [01:37<19:22, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2615/33666 [01:37<19:22, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2615/33666 [01:37<19:22, 26.71it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2616/33666 [01:37<19:22, 26.71it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2616/33666 [01:37<19:22, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2617/33666 [01:37<19:22, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2617/33666 [01:37<19:22, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2618/33666 [01:38<19:22, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2618/33666 [01:38<19:22, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2619/33666 [01:38<19:22, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2619/33666 [01:38<19:22, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2620/33666 [01:38<19:22, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2620/33666 [01:38<19:22, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2621/33666 [01:38<19:22, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2621/33666 [01:38<19:22, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2622/33666 [01:38<19:22, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2622/33666 [01:38<19:22, 26.71it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2623/33666 [01:38<19:22, 26.71it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2623/33666 [01:38<19:22, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2624/33666 [01:38<19:22, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2624/33666 [01:38<19:22, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2625/33666 [01:38<19:22, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2625/33666 [01:38<19:22, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2626/33666 [01:38<19:22, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2626/33666 [01:38<19:22, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2627/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2627/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2628/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2628/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2629/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2629/33666 [01:38<19:21, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2630/33666 [01:38<19:21, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2630/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2631/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2631/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2632/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2632/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2633/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2633/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2634/33666 [01:38<19:21, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2634/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2635/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2635/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2636/33666 [01:38<19:21, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2636/33666 [01:38<19:21, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2637/33666 [01:38<19:21, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2637/33666 [01:38<19:21, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2638/33666 [01:38<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2638/33666 [01:38<19:21, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2639/33666 [01:38<19:21, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2639/33666 [01:38<19:21, 26.72it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2640/33666 [01:38<19:21, 26.72it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2640/33666 [01:38<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2641/33666 [01:38<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2641/33666 [01:38<19:21, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2642/33666 [01:38<19:21, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2642/33666 [01:38<19:21, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2643/33666 [01:38<19:21, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2643/33666 [01:38<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2644/33666 [01:38<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2644/33666 [01:38<19:21, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2645/33666 [01:38<19:21, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2645/33666 [01:38<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2646/33666 [01:39<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2646/33666 [01:39<19:21, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2647/33666 [01:39<19:21, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2647/33666 [01:39<19:21, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2648/33666 [01:39<19:21, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2648/33666 [01:39<19:21, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2649/33666 [01:39<19:21, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2649/33666 [01:39<19:21, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2650/33666 [01:39<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2650/33666 [01:39<19:21, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2651/33666 [01:39<19:21, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2651/33666 [01:39<19:21, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2652/33666 [01:39<19:20, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2652/33666 [01:39<19:20, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2653/33666 [01:39<19:20, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2653/33666 [01:39<19:20, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2654/33666 [01:39<19:20, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2654/33666 [01:39<19:20, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2655/33666 [01:39<19:20, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2655/33666 [01:39<19:20, 26.72it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2656/33666 [01:39<19:20, 26.72it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2656/33666 [01:39<19:20, 26.72it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2657/33666 [01:39<19:20, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2657/33666 [01:39<19:20, 26.71it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2658/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2658/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2659/33666 [01:39<19:20, 26.71it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2659/33666 [01:39<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2660/33666 [01:39<19:20, 26.72it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2660/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2661/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2661/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2662/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2662/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2663/33666 [01:39<19:20, 26.72it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2663/33666 [01:39<19:20, 26.72it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2664/33666 [01:39<19:20, 26.72it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2664/33666 [01:39<19:20, 26.72it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2665/33666 [01:39<19:20, 26.72it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2665/33666 [01:39<19:20, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2666/33666 [01:39<19:20, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2666/33666 [01:39<19:20, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2667/33666 [01:39<19:20, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2667/33666 [01:39<19:20, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2668/33666 [01:39<19:20, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2668/33666 [01:39<19:20, 26.72it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2669/33666 [01:39<19:20, 26.72it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2669/33666 [01:39<19:20, 26.72it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2670/33666 [01:39<19:20, 26.72it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2670/33666 [01:39<19:20, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2671/33666 [01:39<19:20, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2671/33666 [01:39<19:20, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2672/33666 [01:40<19:20, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2672/33666 [01:40<19:20, 26.72it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2673/33666 [01:40<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2673/33666 [01:40<19:20, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2674/33666 [01:40<19:20, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2674/33666 [01:40<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2675/33666 [01:40<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2675/33666 [01:40<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2676/33666 [01:40<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2676/33666 [01:40<19:20, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2677/33666 [01:40<19:19, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2677/33666 [01:40<19:19, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2678/33666 [01:40<19:20, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2678/33666 [01:40<19:20, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2679/33666 [01:40<19:19, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2679/33666 [01:40<19:19, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2680/33666 [01:40<19:19, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2680/33666 [01:40<19:19, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2681/33666 [01:40<19:19, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2681/33666 [01:40<19:19, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2682/33666 [01:40<19:19, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2682/33666 [01:40<19:19, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2683/33666 [01:40<19:19, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2683/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2684/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2684/33666 [01:40<19:19, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2685/33666 [01:40<19:19, 26.71it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2685/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2686/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2686/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2687/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2687/33666 [01:40<19:19, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2688/33666 [01:40<19:19, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2688/33666 [01:40<19:19, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2689/33666 [01:40<19:19, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2689/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2690/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2690/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2691/33666 [01:40<19:19, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2691/33666 [01:40<19:19, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2692/33666 [01:40<19:19, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2692/33666 [01:40<19:19, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2693/33666 [01:40<19:19, 26.71it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2693/33666 [01:40<19:19, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2694/33666 [01:40<19:19, 26.71it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2694/33666 [01:40<19:19, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2695/33666 [01:40<19:19, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2695/33666 [01:40<19:19, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2696/33666 [01:40<19:19, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2696/33666 [01:40<19:19, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2697/33666 [01:40<19:19, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2697/33666 [01:40<19:19, 26.71it/s, loss=4.72, v_num=0]
Epoch 0:   8%|▊         | 2698/33666 [01:41<19:19, 26.71it/s, loss=4.72, v_num=0]
Epoch 0:   8%|▊         | 2698/33666 [01:41<19:19, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2699/33666 [01:41<19:19, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2699/33666 [01:41<19:19, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2700/33666 [01:41<19:19, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2700/33666 [01:41<19:19, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2701/33666 [01:41<19:19, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2701/33666 [01:41<19:19, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2702/33666 [01:41<19:19, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2702/33666 [01:41<19:19, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2703/33666 [01:41<19:19, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2703/33666 [01:41<19:19, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2704/33666 [01:41<19:19, 26.70it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2704/33666 [01:41<19:19, 26.70it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2705/33666 [01:41<19:19, 26.70it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2705/33666 [01:41<19:19, 26.70it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2706/33666 [01:41<19:19, 26.70it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2706/33666 [01:41<19:19, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2707/33666 [01:41<19:19, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2707/33666 [01:41<19:19, 26.70it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2708/33666 [01:41<19:19, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2708/33666 [01:41<19:19, 26.71it/s, loss=4.75, v_num=0]
Epoch 0:   8%|▊         | 2709/33666 [01:41<19:19, 26.71it/s, loss=4.75, v_num=0]
Epoch 0:   8%|▊         | 2709/33666 [01:41<19:19, 26.71it/s, loss=4.74, v_num=0]
Epoch 0:   8%|▊         | 2710/33666 [01:41<19:19, 26.70it/s, loss=4.74, v_num=0]
Epoch 0:   8%|▊         | 2710/33666 [01:41<19:19, 26.70it/s, loss=4.74, v_num=0]
Epoch 0:   8%|▊         | 2711/33666 [01:41<19:19, 26.71it/s, loss=4.74, v_num=0]
Epoch 0:   8%|▊         | 2711/33666 [01:41<19:19, 26.71it/s, loss=4.74, v_num=0]
Epoch 0:   8%|▊         | 2712/33666 [01:41<19:19, 26.71it/s, loss=4.74, v_num=0]
Epoch 0:   8%|▊         | 2712/33666 [01:41<19:19, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2713/33666 [01:41<19:19, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2713/33666 [01:41<19:19, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2714/33666 [01:41<19:18, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2714/33666 [01:41<19:18, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2715/33666 [01:41<19:18, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2715/33666 [01:41<19:18, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2716/33666 [01:41<19:18, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2716/33666 [01:41<19:18, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2717/33666 [01:41<19:18, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2717/33666 [01:41<19:18, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2718/33666 [01:41<19:18, 26.71it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2718/33666 [01:41<19:18, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2719/33666 [01:41<19:18, 26.71it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2719/33666 [01:41<19:18, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2720/33666 [01:41<19:18, 26.71it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2720/33666 [01:41<19:18, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2721/33666 [01:41<19:18, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2721/33666 [01:41<19:18, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2722/33666 [01:41<19:18, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2722/33666 [01:41<19:18, 26.71it/s, loss=4.72, v_num=0]
Epoch 0:   8%|▊         | 2723/33666 [01:41<19:18, 26.70it/s, loss=4.72, v_num=0]
Epoch 0:   8%|▊         | 2723/33666 [01:41<19:18, 26.70it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2724/33666 [01:42<19:18, 26.71it/s, loss=4.71, v_num=0]
Epoch 0:   8%|▊         | 2724/33666 [01:42<19:18, 26.71it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2725/33666 [01:42<19:18, 26.70it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2725/33666 [01:42<19:18, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2726/33666 [01:42<19:18, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2726/33666 [01:42<19:18, 26.70it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2727/33666 [01:42<19:18, 26.70it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2727/33666 [01:42<19:18, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2728/33666 [01:42<19:18, 26.70it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2728/33666 [01:42<19:18, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2729/33666 [01:42<19:18, 26.70it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2729/33666 [01:42<19:18, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2730/33666 [01:42<19:18, 26.70it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2730/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2731/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2731/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2732/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2732/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2733/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2733/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2734/33666 [01:42<19:18, 26.70it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2734/33666 [01:42<19:18, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2735/33666 [01:42<19:18, 26.70it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2735/33666 [01:42<19:18, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2736/33666 [01:42<19:18, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2736/33666 [01:42<19:18, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2737/33666 [01:42<19:18, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2737/33666 [01:42<19:18, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2738/33666 [01:42<19:18, 26.70it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2738/33666 [01:42<19:18, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2739/33666 [01:42<19:18, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2739/33666 [01:42<19:18, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2740/33666 [01:42<19:18, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2740/33666 [01:42<19:18, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2741/33666 [01:42<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2741/33666 [01:42<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2742/33666 [01:42<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2742/33666 [01:42<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2743/33666 [01:42<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2743/33666 [01:42<19:17, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2744/33666 [01:42<19:17, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2744/33666 [01:42<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2745/33666 [01:42<19:17, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2745/33666 [01:42<19:17, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2746/33666 [01:42<19:17, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2746/33666 [01:42<19:17, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2747/33666 [01:42<19:17, 26.70it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2747/33666 [01:42<19:17, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2748/33666 [01:42<19:17, 26.70it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2748/33666 [01:42<19:17, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2749/33666 [01:42<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2749/33666 [01:42<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2750/33666 [01:42<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2750/33666 [01:42<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2751/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2751/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2752/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2752/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2753/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2753/33666 [01:43<19:17, 26.70it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2754/33666 [01:43<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2754/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2755/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2755/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2756/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2756/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2757/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2757/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2758/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2758/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2759/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2759/33666 [01:43<19:17, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2760/33666 [01:43<19:17, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2760/33666 [01:43<19:17, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2761/33666 [01:43<19:17, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2761/33666 [01:43<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2762/33666 [01:43<19:17, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2762/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2763/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2763/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2764/33666 [01:43<19:17, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2764/33666 [01:43<19:17, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2765/33666 [01:43<19:17, 26.71it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2765/33666 [01:43<19:17, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2766/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2766/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2767/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2767/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2768/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2768/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2769/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2769/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2770/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2770/33666 [01:43<19:16, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2771/33666 [01:43<19:16, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2771/33666 [01:43<19:16, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2772/33666 [01:43<19:16, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2772/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2773/33666 [01:43<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2773/33666 [01:43<19:16, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2774/33666 [01:43<19:16, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2774/33666 [01:43<19:16, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2775/33666 [01:43<19:16, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2775/33666 [01:43<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2776/33666 [01:43<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2776/33666 [01:43<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2777/33666 [01:43<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2777/33666 [01:43<19:16, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2778/33666 [01:44<19:16, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2778/33666 [01:44<19:16, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2779/33666 [01:44<19:16, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2779/33666 [01:44<19:16, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2780/33666 [01:44<19:16, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2780/33666 [01:44<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2781/33666 [01:44<19:16, 26.71it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2781/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2782/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2782/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2783/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2783/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2784/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2784/33666 [01:44<19:16, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2785/33666 [01:44<19:16, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2785/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2786/33666 [01:44<19:16, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2786/33666 [01:44<19:16, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2787/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2787/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2788/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2788/33666 [01:44<19:15, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2789/33666 [01:44<19:15, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2789/33666 [01:44<19:15, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2790/33666 [01:44<19:15, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2790/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2791/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2791/33666 [01:44<19:15, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2792/33666 [01:44<19:15, 26.71it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2792/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2793/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2793/33666 [01:44<19:15, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2794/33666 [01:44<19:15, 26.71it/s, loss=4.54, v_num=0]
Epoch 0:   8%|▊         | 2794/33666 [01:44<19:15, 26.71it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2795/33666 [01:44<19:15, 26.71it/s, loss=4.53, v_num=0]
Epoch 0:   8%|▊         | 2795/33666 [01:44<19:15, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2796/33666 [01:44<19:15, 26.71it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2796/33666 [01:44<19:15, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2797/33666 [01:44<19:15, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2797/33666 [01:44<19:15, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2798/33666 [01:44<19:15, 26.71it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2798/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2799/33666 [01:44<19:15, 26.71it/s, loss=4.57, v_num=0]
Epoch 0:   8%|▊         | 2799/33666 [01:44<19:15, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2800/33666 [01:44<19:15, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2800/33666 [01:44<19:15, 26.71it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2801/33666 [01:44<19:15, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2801/33666 [01:44<19:15, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2802/33666 [01:44<19:15, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2802/33666 [01:44<19:15, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2803/33666 [01:44<19:15, 26.72it/s, loss=4.56, v_num=0]
Epoch 0:   8%|▊         | 2803/33666 [01:44<19:15, 26.72it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2804/33666 [01:44<19:15, 26.72it/s, loss=4.55, v_num=0]
Epoch 0:   8%|▊         | 2804/33666 [01:44<19:15, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2805/33666 [01:44<19:15, 26.72it/s, loss=4.58, v_num=0]
Epoch 0:   8%|▊         | 2805/33666 [01:44<19:15, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2806/33666 [01:45<19:15, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2806/33666 [01:45<19:15, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2807/33666 [01:45<19:15, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2807/33666 [01:45<19:15, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2808/33666 [01:45<19:14, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2808/33666 [01:45<19:14, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2809/33666 [01:45<19:14, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2809/33666 [01:45<19:14, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2810/33666 [01:45<19:14, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2810/33666 [01:45<19:14, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2811/33666 [01:45<19:14, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2811/33666 [01:45<19:14, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2812/33666 [01:45<19:14, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2812/33666 [01:45<19:14, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2813/33666 [01:45<19:14, 26.72it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2813/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2814/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2814/33666 [01:45<19:14, 26.72it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2815/33666 [01:45<19:14, 26.72it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2815/33666 [01:45<19:14, 26.72it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2816/33666 [01:45<19:14, 26.72it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2816/33666 [01:45<19:14, 26.72it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2817/33666 [01:45<19:14, 26.72it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2817/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2818/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2818/33666 [01:45<19:14, 26.72it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2819/33666 [01:45<19:14, 26.72it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2819/33666 [01:45<19:14, 26.72it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2820/33666 [01:45<19:14, 26.72it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2820/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2821/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2821/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2822/33666 [01:45<19:14, 26.72it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2822/33666 [01:45<19:14, 26.72it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2823/33666 [01:45<19:14, 26.72it/s, loss=4.68, v_num=0]
Epoch 0:   8%|▊         | 2823/33666 [01:45<19:14, 26.72it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2824/33666 [01:45<19:14, 26.72it/s, loss=4.7, v_num=0]
Epoch 0:   8%|▊         | 2824/33666 [01:45<19:14, 26.72it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2825/33666 [01:45<19:14, 26.72it/s, loss=4.69, v_num=0]
Epoch 0:   8%|▊         | 2825/33666 [01:45<19:14, 26.72it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2826/33666 [01:45<19:14, 26.72it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2826/33666 [01:45<19:14, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2827/33666 [01:45<19:14, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2827/33666 [01:45<19:14, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2828/33666 [01:45<19:13, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2828/33666 [01:45<19:13, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2829/33666 [01:45<19:13, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2829/33666 [01:45<19:13, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2830/33666 [01:45<19:13, 26.72it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2830/33666 [01:45<19:13, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2831/33666 [01:45<19:13, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2831/33666 [01:45<19:13, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2832/33666 [01:45<19:13, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2832/33666 [01:45<19:13, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2833/33666 [01:46<19:13, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2833/33666 [01:46<19:13, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2834/33666 [01:46<19:13, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2834/33666 [01:46<19:13, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2835/33666 [01:46<19:13, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2835/33666 [01:46<19:13, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2836/33666 [01:46<19:13, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2836/33666 [01:46<19:13, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2837/33666 [01:46<19:13, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2837/33666 [01:46<19:13, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2838/33666 [01:46<19:13, 26.72it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2838/33666 [01:46<19:13, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2839/33666 [01:46<19:13, 26.72it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2839/33666 [01:46<19:13, 26.72it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2840/33666 [01:46<19:13, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2840/33666 [01:46<19:13, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2841/33666 [01:46<19:13, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2841/33666 [01:46<19:13, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2842/33666 [01:46<19:13, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2842/33666 [01:46<19:13, 26.72it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2843/33666 [01:46<19:13, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2843/33666 [01:46<19:13, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2844/33666 [01:46<19:13, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2844/33666 [01:46<19:13, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2845/33666 [01:46<19:13, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2845/33666 [01:46<19:13, 26.73it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2846/33666 [01:46<19:13, 26.73it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2846/33666 [01:46<19:13, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2847/33666 [01:46<19:13, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2847/33666 [01:46<19:13, 26.73it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2848/33666 [01:46<19:13, 26.73it/s, loss=4.66, v_num=0]
Epoch 0:   8%|▊         | 2848/33666 [01:46<19:13, 26.73it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2849/33666 [01:46<19:13, 26.73it/s, loss=4.67, v_num=0]
Epoch 0:   8%|▊         | 2849/33666 [01:46<19:13, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2850/33666 [01:46<19:12, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2850/33666 [01:46<19:13, 26.73it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2851/33666 [01:46<19:12, 26.73it/s, loss=4.64, v_num=0]
Epoch 0:   8%|▊         | 2851/33666 [01:46<19:12, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2852/33666 [01:46<19:12, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   8%|▊         | 2852/33666 [01:46<19:12, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2853/33666 [01:46<19:12, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   8%|▊         | 2853/33666 [01:46<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2854/33666 [01:46<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2854/33666 [01:46<19:12, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2855/33666 [01:46<19:12, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   8%|▊         | 2855/33666 [01:46<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2856/33666 [01:46<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2856/33666 [01:46<19:12, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2857/33666 [01:46<19:12, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   8%|▊         | 2857/33666 [01:46<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2858/33666 [01:46<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2858/33666 [01:46<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2859/33666 [01:46<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   8%|▊         | 2859/33666 [01:46<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2860/33666 [01:46<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2860/33666 [01:46<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2861/33666 [01:47<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   8%|▊         | 2861/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2862/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2862/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2863/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2863/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2864/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2864/33666 [01:47<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2865/33666 [01:47<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2865/33666 [01:47<19:12, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2866/33666 [01:47<19:12, 26.73it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2866/33666 [01:47<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2867/33666 [01:47<19:12, 26.73it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2867/33666 [01:47<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2868/33666 [01:47<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2868/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2869/33666 [01:47<19:12, 26.73it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2869/33666 [01:47<19:12, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2870/33666 [01:47<19:11, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2870/33666 [01:47<19:11, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2871/33666 [01:47<19:11, 26.73it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2871/33666 [01:47<19:11, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2872/33666 [01:47<19:11, 26.73it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2872/33666 [01:47<19:11, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2873/33666 [01:47<19:11, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2873/33666 [01:47<19:11, 26.73it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2874/33666 [01:47<19:11, 26.73it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2874/33666 [01:47<19:11, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2875/33666 [01:47<19:11, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2875/33666 [01:47<19:11, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2876/33666 [01:47<19:11, 26.73it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2876/33666 [01:47<19:11, 26.73it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2877/33666 [01:47<19:11, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2877/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2878/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2878/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2879/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2879/33666 [01:47<19:11, 26.74it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▊         | 2880/33666 [01:47<19:11, 26.74it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▊         | 2880/33666 [01:47<19:11, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2881/33666 [01:47<19:11, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2881/33666 [01:47<19:11, 26.74it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▊         | 2882/33666 [01:47<19:11, 26.74it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▊         | 2882/33666 [01:47<19:11, 26.74it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▊         | 2883/33666 [01:47<19:11, 26.74it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▊         | 2883/33666 [01:47<19:11, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2884/33666 [01:47<19:11, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2884/33666 [01:47<19:11, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2885/33666 [01:47<19:11, 26.74it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2885/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2886/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2886/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2887/33666 [01:47<19:11, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2887/33666 [01:47<19:11, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2888/33666 [01:48<19:11, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2888/33666 [01:48<19:11, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2889/33666 [01:48<19:10, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2889/33666 [01:48<19:10, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2890/33666 [01:48<19:10, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2890/33666 [01:48<19:10, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2891/33666 [01:48<19:10, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2891/33666 [01:48<19:10, 26.74it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2892/33666 [01:48<19:10, 26.74it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2892/33666 [01:48<19:10, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2893/33666 [01:48<19:10, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2893/33666 [01:48<19:10, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2894/33666 [01:48<19:10, 26.74it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2894/33666 [01:48<19:10, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2895/33666 [01:48<19:10, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2895/33666 [01:48<19:10, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2896/33666 [01:48<19:10, 26.74it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▊         | 2896/33666 [01:48<19:10, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2897/33666 [01:48<19:10, 26.74it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2897/33666 [01:48<19:10, 26.74it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2898/33666 [01:48<19:10, 26.74it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2898/33666 [01:48<19:10, 26.74it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2899/33666 [01:48<19:10, 26.74it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2899/33666 [01:48<19:10, 26.74it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2900/33666 [01:48<19:10, 26.74it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2900/33666 [01:48<19:10, 26.74it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2901/33666 [01:48<19:10, 26.74it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2901/33666 [01:48<19:10, 26.74it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▊         | 2902/33666 [01:48<19:10, 26.74it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▊         | 2902/33666 [01:48<19:10, 26.74it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▊         | 2903/33666 [01:48<19:10, 26.74it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▊         | 2903/33666 [01:48<19:10, 26.74it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▊         | 2904/33666 [01:48<19:10, 26.74it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▊         | 2904/33666 [01:48<19:10, 26.74it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▊         | 2905/33666 [01:48<19:10, 26.74it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▊         | 2905/33666 [01:48<19:10, 26.74it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▊         | 2906/33666 [01:48<19:10, 26.74it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▊         | 2906/33666 [01:48<19:10, 26.74it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▊         | 2907/33666 [01:48<19:10, 26.74it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▊         | 2907/33666 [01:48<19:10, 26.74it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▊         | 2908/33666 [01:48<19:10, 26.74it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▊         | 2908/33666 [01:48<19:10, 26.74it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▊         | 2909/33666 [01:48<19:10, 26.75it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▊         | 2909/33666 [01:48<19:10, 26.74it/s, loss=4.51, v_num=0]
Epoch 0:   9%|▊         | 2910/33666 [01:48<19:09, 26.75it/s, loss=4.51, v_num=0]
Epoch 0:   9%|▊         | 2910/33666 [01:48<19:09, 26.75it/s, loss=4.49, v_num=0]
Epoch 0:   9%|▊         | 2911/33666 [01:48<19:09, 26.75it/s, loss=4.49, v_num=0]
Epoch 0:   9%|▊         | 2911/33666 [01:48<19:09, 26.75it/s, loss=4.47, v_num=0]
Epoch 0:   9%|▊         | 2912/33666 [01:48<19:09, 26.75it/s, loss=4.47, v_num=0]
Epoch 0:   9%|▊         | 2912/33666 [01:48<19:09, 26.75it/s, loss=4.5, v_num=0]
Epoch 0:   9%|▊         | 2913/33666 [01:48<19:09, 26.75it/s, loss=4.5, v_num=0]
Epoch 0:   9%|▊         | 2913/33666 [01:48<19:09, 26.75it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▊         | 2914/33666 [01:48<19:09, 26.75it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▊         | 2914/33666 [01:48<19:09, 26.75it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▊         | 2915/33666 [01:48<19:09, 26.75it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▊         | 2915/33666 [01:48<19:09, 26.75it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2916/33666 [01:49<19:09, 26.75it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▊         | 2916/33666 [01:49<19:09, 26.75it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▊         | 2917/33666 [01:49<19:09, 26.75it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▊         | 2917/33666 [01:49<19:09, 26.75it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▊         | 2918/33666 [01:49<19:09, 26.75it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▊         | 2918/33666 [01:49<19:09, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2919/33666 [01:49<19:09, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2919/33666 [01:49<19:09, 26.75it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2920/33666 [01:49<19:09, 26.75it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▊         | 2920/33666 [01:49<19:09, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2921/33666 [01:49<19:09, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2921/33666 [01:49<19:09, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2922/33666 [01:49<19:09, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2922/33666 [01:49<19:09, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2923/33666 [01:49<19:09, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2923/33666 [01:49<19:09, 26.75it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2924/33666 [01:49<19:09, 26.75it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▊         | 2924/33666 [01:49<19:09, 26.75it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2925/33666 [01:49<19:09, 26.75it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▊         | 2925/33666 [01:49<19:09, 26.75it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▊         | 2926/33666 [01:49<19:09, 26.75it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▊         | 2926/33666 [01:49<19:09, 26.75it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▊         | 2927/33666 [01:49<19:09, 26.75it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▊         | 2927/33666 [01:49<19:09, 26.75it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▊         | 2928/33666 [01:49<19:09, 26.75it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▊         | 2928/33666 [01:49<19:09, 26.75it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2929/33666 [01:49<19:09, 26.75it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2929/33666 [01:49<19:09, 26.75it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▊         | 2930/33666 [01:49<19:09, 26.75it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▊         | 2930/33666 [01:49<19:09, 26.75it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▊         | 2931/33666 [01:49<19:08, 26.75it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▊         | 2931/33666 [01:49<19:08, 26.75it/s, loss=4.72, v_num=0]
Epoch 0:   9%|▊         | 2932/33666 [01:49<19:08, 26.75it/s, loss=4.72, v_num=0]
Epoch 0:   9%|▊         | 2932/33666 [01:49<19:08, 26.75it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2933/33666 [01:49<19:08, 26.75it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▊         | 2933/33666 [01:49<19:08, 26.75it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2934/33666 [01:49<19:08, 26.75it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▊         | 2934/33666 [01:49<19:08, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2935/33666 [01:49<19:08, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2935/33666 [01:49<19:08, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2936/33666 [01:49<19:08, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2936/33666 [01:49<19:08, 26.75it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2937/33666 [01:49<19:08, 26.75it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2937/33666 [01:49<19:08, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2938/33666 [01:49<19:08, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2938/33666 [01:49<19:08, 26.75it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2939/33666 [01:49<19:08, 26.75it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2939/33666 [01:49<19:08, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2940/33666 [01:49<19:08, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2940/33666 [01:49<19:08, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2941/33666 [01:49<19:08, 26.75it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▊         | 2941/33666 [01:49<19:08, 26.75it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2942/33666 [01:49<19:08, 26.75it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▊         | 2942/33666 [01:49<19:08, 26.75it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2943/33666 [01:49<19:08, 26.76it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2943/33666 [01:49<19:08, 26.76it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2944/33666 [01:50<19:08, 26.76it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2944/33666 [01:50<19:08, 26.76it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2945/33666 [01:50<19:08, 26.76it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▊         | 2945/33666 [01:50<19:08, 26.76it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 2946/33666 [01:50<19:07, 26.76it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 2946/33666 [01:50<19:07, 26.76it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 2947/33666 [01:50<19:07, 26.76it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 2947/33666 [01:50<19:07, 26.76it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 2948/33666 [01:50<19:07, 26.76it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 2948/33666 [01:50<19:07, 26.76it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 2949/33666 [01:50<19:07, 26.76it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 2949/33666 [01:50<19:07, 26.76it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 2950/33666 [01:50<19:07, 26.76it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 2950/33666 [01:50<19:07, 26.76it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 2951/33666 [01:50<19:07, 26.76it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 2951/33666 [01:50<19:07, 26.76it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 2952/33666 [01:50<19:07, 26.76it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 2952/33666 [01:50<19:07, 26.76it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2953/33666 [01:50<19:07, 26.76it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2953/33666 [01:50<19:07, 26.76it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2954/33666 [01:50<19:07, 26.76it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2954/33666 [01:50<19:07, 26.76it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 2955/33666 [01:50<19:07, 26.76it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 2955/33666 [01:50<19:07, 26.76it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2956/33666 [01:50<19:07, 26.76it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2956/33666 [01:50<19:07, 26.76it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2957/33666 [01:50<19:07, 26.77it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2957/33666 [01:50<19:07, 26.77it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2958/33666 [01:50<19:07, 26.77it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 2958/33666 [01:50<19:07, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 2959/33666 [01:50<19:07, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 2959/33666 [01:50<19:07, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2960/33666 [01:50<19:07, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2960/33666 [01:50<19:07, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2961/33666 [01:50<19:07, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2961/33666 [01:50<19:07, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 2962/33666 [01:50<19:07, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 2962/33666 [01:50<19:07, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2963/33666 [01:50<19:07, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 2963/33666 [01:50<19:07, 26.77it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 2964/33666 [01:50<19:06, 26.77it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 2964/33666 [01:50<19:06, 26.77it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 2965/33666 [01:50<19:07, 26.77it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 2965/33666 [01:50<19:07, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 2966/33666 [01:50<19:06, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 2966/33666 [01:50<19:06, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 2967/33666 [01:50<19:06, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 2967/33666 [01:50<19:06, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 2968/33666 [01:50<19:06, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 2968/33666 [01:50<19:06, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 2969/33666 [01:50<19:06, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 2969/33666 [01:50<19:06, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 2970/33666 [01:50<19:06, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 2970/33666 [01:50<19:06, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2971/33666 [01:50<19:06, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2971/33666 [01:50<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2972/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2972/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2973/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2973/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2974/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2974/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2975/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2975/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2976/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2976/33666 [01:51<19:06, 26.77it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 2977/33666 [01:51<19:06, 26.77it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 2977/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2978/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2978/33666 [01:51<19:06, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2979/33666 [01:51<19:06, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2979/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2980/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2980/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2981/33666 [01:51<19:06, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2981/33666 [01:51<19:06, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2982/33666 [01:51<19:06, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2982/33666 [01:51<19:06, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2983/33666 [01:51<19:06, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2983/33666 [01:51<19:06, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2984/33666 [01:51<19:06, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2984/33666 [01:51<19:06, 26.77it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 2985/33666 [01:51<19:06, 26.77it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 2985/33666 [01:51<19:06, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2986/33666 [01:51<19:05, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2986/33666 [01:51<19:05, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2987/33666 [01:51<19:05, 26.77it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 2987/33666 [01:51<19:05, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2988/33666 [01:51<19:05, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2988/33666 [01:51<19:05, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2989/33666 [01:51<19:05, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2989/33666 [01:51<19:05, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2990/33666 [01:51<19:05, 26.77it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 2990/33666 [01:51<19:05, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2991/33666 [01:51<19:05, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2991/33666 [01:51<19:05, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 2992/33666 [01:51<19:05, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 2992/33666 [01:51<19:05, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 2993/33666 [01:51<19:05, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 2993/33666 [01:51<19:05, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2994/33666 [01:51<19:05, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 2994/33666 [01:51<19:05, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 2995/33666 [01:51<19:05, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 2995/33666 [01:51<19:05, 26.77it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 2996/33666 [01:51<19:05, 26.77it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 2996/33666 [01:51<19:05, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 2997/33666 [01:51<19:05, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 2997/33666 [01:51<19:05, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 2998/33666 [01:51<19:05, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 2998/33666 [01:51<19:05, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 2999/33666 [01:52<19:05, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 2999/33666 [01:52<19:05, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3000/33666 [01:52<19:05, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3000/33666 [01:52<19:05, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3001/33666 [01:52<19:05, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3001/33666 [01:52<19:05, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3002/33666 [01:52<19:05, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3002/33666 [01:52<19:05, 26.78it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3003/33666 [01:52<19:05, 26.78it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3003/33666 [01:52<19:05, 26.78it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3004/33666 [01:52<19:05, 26.78it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3004/33666 [01:52<19:05, 26.78it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 3005/33666 [01:52<19:05, 26.78it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 3005/33666 [01:52<19:05, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3006/33666 [01:52<19:05, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3006/33666 [01:52<19:05, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3007/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3007/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3008/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3008/33666 [01:52<19:04, 26.78it/s, loss=4.73, v_num=0]
Epoch 0:   9%|▉         | 3009/33666 [01:52<19:04, 26.78it/s, loss=4.73, v_num=0]
Epoch 0:   9%|▉         | 3009/33666 [01:52<19:04, 26.78it/s, loss=4.73, v_num=0]
Epoch 0:   9%|▉         | 3010/33666 [01:52<19:04, 26.78it/s, loss=4.73, v_num=0]
Epoch 0:   9%|▉         | 3010/33666 [01:52<19:04, 26.78it/s, loss=4.75, v_num=0]
Epoch 0:   9%|▉         | 3011/33666 [01:52<19:04, 26.78it/s, loss=4.75, v_num=0]
Epoch 0:   9%|▉         | 3011/33666 [01:52<19:04, 26.78it/s, loss=4.75, v_num=0]
Epoch 0:   9%|▉         | 3012/33666 [01:52<19:04, 26.78it/s, loss=4.75, v_num=0]
Epoch 0:   9%|▉         | 3012/33666 [01:52<19:04, 26.78it/s, loss=4.74, v_num=0]
Epoch 0:   9%|▉         | 3013/33666 [01:52<19:04, 26.78it/s, loss=4.74, v_num=0]
Epoch 0:   9%|▉         | 3013/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3014/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3014/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3015/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3015/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3016/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3016/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3017/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3017/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3018/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3018/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3019/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3019/33666 [01:52<19:04, 26.78it/s, loss=4.72, v_num=0]
Epoch 0:   9%|▉         | 3020/33666 [01:52<19:04, 26.78it/s, loss=4.72, v_num=0]
Epoch 0:   9%|▉         | 3020/33666 [01:52<19:04, 26.78it/s, loss=4.73, v_num=0]
Epoch 0:   9%|▉         | 3021/33666 [01:52<19:04, 26.78it/s, loss=4.73, v_num=0]
Epoch 0:   9%|▉         | 3021/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3022/33666 [01:52<19:04, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3022/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3023/33666 [01:52<19:04, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3023/33666 [01:52<19:04, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3024/33666 [01:52<19:04, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3024/33666 [01:52<19:04, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3025/33666 [01:52<19:04, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3025/33666 [01:52<19:04, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3026/33666 [01:52<19:04, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3026/33666 [01:52<19:04, 26.78it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3027/33666 [01:53<19:04, 26.78it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3027/33666 [01:53<19:04, 26.78it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3028/33666 [01:53<19:04, 26.78it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3028/33666 [01:53<19:04, 26.78it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3029/33666 [01:53<19:04, 26.78it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3029/33666 [01:53<19:04, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3030/33666 [01:53<19:03, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3030/33666 [01:53<19:03, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3031/33666 [01:53<19:03, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3031/33666 [01:53<19:03, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3032/33666 [01:53<19:03, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3032/33666 [01:53<19:03, 26.78it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3033/33666 [01:53<19:03, 26.78it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3033/33666 [01:53<19:03, 26.78it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3034/33666 [01:53<19:03, 26.78it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3034/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3035/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3035/33666 [01:53<19:03, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3036/33666 [01:53<19:03, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3036/33666 [01:53<19:03, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3037/33666 [01:53<19:03, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3037/33666 [01:53<19:03, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3038/33666 [01:53<19:03, 26.78it/s, loss=4.7, v_num=0]
Epoch 0:   9%|▉         | 3038/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3039/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3039/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3040/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3040/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3041/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3041/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3042/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3042/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3043/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3043/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3044/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3044/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3045/33666 [01:53<19:03, 26.78it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3045/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3046/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3046/33666 [01:53<19:03, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3047/33666 [01:53<19:03, 26.78it/s, loss=4.71, v_num=0]
Epoch 0:   9%|▉         | 3047/33666 [01:53<19:03, 26.78it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3048/33666 [01:53<19:03, 26.79it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3048/33666 [01:53<19:03, 26.79it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3049/33666 [01:53<19:03, 26.79it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3049/33666 [01:53<19:03, 26.79it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3050/33666 [01:53<19:02, 26.79it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3050/33666 [01:53<19:03, 26.79it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3051/33666 [01:53<19:02, 26.79it/s, loss=4.68, v_num=0]
Epoch 0:   9%|▉         | 3051/33666 [01:53<19:02, 26.79it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3052/33666 [01:53<19:02, 26.79it/s, loss=4.69, v_num=0]
Epoch 0:   9%|▉         | 3052/33666 [01:53<19:02, 26.79it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 3053/33666 [01:53<19:02, 26.79it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 3053/33666 [01:53<19:02, 26.79it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 3054/33666 [01:54<19:02, 26.79it/s, loss=4.67, v_num=0]
Epoch 0:   9%|▉         | 3054/33666 [01:54<19:02, 26.79it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3055/33666 [01:54<19:02, 26.79it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3055/33666 [01:54<19:02, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3056/33666 [01:54<19:02, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3056/33666 [01:54<19:02, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3057/33666 [01:54<19:02, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3057/33666 [01:54<19:02, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3058/33666 [01:54<19:02, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3058/33666 [01:54<19:02, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3059/33666 [01:54<19:02, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3059/33666 [01:54<19:02, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3060/33666 [01:54<19:02, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3060/33666 [01:54<19:02, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3061/33666 [01:54<19:02, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3061/33666 [01:54<19:02, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3062/33666 [01:54<19:02, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3062/33666 [01:54<19:02, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3063/33666 [01:54<19:02, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3063/33666 [01:54<19:02, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3064/33666 [01:54<19:02, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3064/33666 [01:54<19:02, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3065/33666 [01:54<19:02, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3065/33666 [01:54<19:02, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3066/33666 [01:54<19:02, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3066/33666 [01:54<19:02, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3067/33666 [01:54<19:02, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3067/33666 [01:54<19:02, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3068/33666 [01:54<19:02, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3068/33666 [01:54<19:02, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3069/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3069/33666 [01:54<19:02, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3070/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3070/33666 [01:54<19:01, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3071/33666 [01:54<19:01, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3071/33666 [01:54<19:01, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3072/33666 [01:54<19:01, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3072/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3073/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3073/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3074/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3074/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3075/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3075/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3076/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3076/33666 [01:54<19:01, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3077/33666 [01:54<19:01, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3077/33666 [01:54<19:01, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3078/33666 [01:54<19:01, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3078/33666 [01:54<19:01, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3079/33666 [01:54<19:01, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3079/33666 [01:54<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3080/33666 [01:54<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3080/33666 [01:54<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3081/33666 [01:54<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3081/33666 [01:54<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3082/33666 [01:55<19:01, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3082/33666 [01:55<19:01, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3083/33666 [01:55<19:01, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3083/33666 [01:55<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3084/33666 [01:55<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3084/33666 [01:55<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3085/33666 [01:55<19:01, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3085/33666 [01:55<19:01, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3086/33666 [01:55<19:01, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3086/33666 [01:55<19:01, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3087/33666 [01:55<19:01, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3087/33666 [01:55<19:01, 26.79it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3088/33666 [01:55<19:01, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3088/33666 [01:55<19:01, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3089/33666 [01:55<19:01, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3089/33666 [01:55<19:01, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3090/33666 [01:55<19:01, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3090/33666 [01:55<19:01, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3091/33666 [01:55<19:01, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3091/33666 [01:55<19:01, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3092/33666 [01:55<19:00, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3092/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3093/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3093/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3094/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3094/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3095/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3095/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3096/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3096/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3097/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3097/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3098/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3098/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3099/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3099/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3100/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3100/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3101/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3101/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3102/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3102/33666 [01:55<19:00, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3103/33666 [01:55<19:00, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3103/33666 [01:55<19:00, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3104/33666 [01:55<19:00, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3104/33666 [01:55<19:00, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3105/33666 [01:55<19:00, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3105/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3106/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3106/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3107/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3107/33666 [01:55<19:00, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3108/33666 [01:55<19:00, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3108/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3109/33666 [01:55<19:00, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3109/33666 [01:55<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3110/33666 [01:56<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3110/33666 [01:56<19:00, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3111/33666 [01:56<18:59, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3111/33666 [01:56<18:59, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3112/33666 [01:56<18:59, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3112/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3113/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3113/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3114/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3114/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3115/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3115/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3116/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3116/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3117/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3117/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3118/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3118/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3119/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3119/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3120/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3120/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3121/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3121/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3122/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3122/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3123/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3123/33666 [01:56<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3124/33666 [01:56<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3124/33666 [01:56<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3125/33666 [01:56<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3125/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3126/33666 [01:56<18:59, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3126/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3127/33666 [01:56<18:59, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:   9%|▉         | 3127/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3128/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3128/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3129/33666 [01:56<18:59, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3129/33666 [01:56<18:59, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3130/33666 [01:56<18:59, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3130/33666 [01:56<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3131/33666 [01:56<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3131/33666 [01:56<18:59, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3132/33666 [01:56<18:59, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3132/33666 [01:56<18:59, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3133/33666 [01:56<18:59, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3133/33666 [01:56<18:59, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3134/33666 [01:56<18:59, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3134/33666 [01:56<18:59, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3135/33666 [01:56<18:59, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3135/33666 [01:56<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3136/33666 [01:57<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3136/33666 [01:57<18:59, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3137/33666 [01:57<18:59, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3137/33666 [01:57<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3138/33666 [01:57<18:59, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3138/33666 [01:57<18:59, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3139/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3139/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3140/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3140/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3141/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3141/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3142/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3142/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3143/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3143/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3144/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3144/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3145/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3145/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3146/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3146/33666 [01:57<18:58, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3147/33666 [01:57<18:58, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3147/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3148/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3148/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3149/33666 [01:57<18:58, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3149/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3150/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3150/33666 [01:57<18:58, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3151/33666 [01:57<18:58, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3151/33666 [01:57<18:58, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3152/33666 [01:57<18:58, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:   9%|▉         | 3152/33666 [01:57<18:58, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3153/33666 [01:57<18:58, 26.80it/s, loss=4.64, v_num=0]
Epoch 0:   9%|▉         | 3153/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3154/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3154/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3155/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3155/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3156/33666 [01:57<18:58, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:   9%|▉         | 3156/33666 [01:57<18:58, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3157/33666 [01:57<18:58, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3157/33666 [01:57<18:58, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3158/33666 [01:57<18:58, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3158/33666 [01:57<18:58, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3159/33666 [01:57<18:58, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3159/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3160/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3160/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3161/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3161/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3162/33666 [01:57<18:58, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3162/33666 [01:57<18:58, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3163/33666 [01:58<18:58, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:   9%|▉         | 3163/33666 [01:58<18:58, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3164/33666 [01:58<18:57, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3164/33666 [01:58<18:57, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3165/33666 [01:58<18:57, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3165/33666 [01:58<18:57, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3166/33666 [01:58<18:57, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3166/33666 [01:58<18:57, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3167/33666 [01:58<18:57, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:   9%|▉         | 3167/33666 [01:58<18:57, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3168/33666 [01:58<18:57, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3168/33666 [01:58<18:57, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3169/33666 [01:58<18:57, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3169/33666 [01:58<18:57, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3170/33666 [01:58<18:57, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3170/33666 [01:58<18:57, 26.81it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 3171/33666 [01:58<18:57, 26.81it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 3171/33666 [01:58<18:57, 26.81it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▉         | 3172/33666 [01:58<18:57, 26.81it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▉         | 3172/33666 [01:58<18:57, 26.81it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▉         | 3173/33666 [01:58<18:57, 26.81it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▉         | 3173/33666 [01:58<18:57, 26.81it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 3174/33666 [01:58<18:57, 26.81it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 3174/33666 [01:58<18:57, 26.81it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 3175/33666 [01:58<18:57, 26.81it/s, loss=4.54, v_num=0]
Epoch 0:   9%|▉         | 3175/33666 [01:58<18:57, 26.81it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 3176/33666 [01:58<18:57, 26.81it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 3176/33666 [01:58<18:57, 26.81it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▉         | 3177/33666 [01:58<18:57, 26.81it/s, loss=4.52, v_num=0]
Epoch 0:   9%|▉         | 3177/33666 [01:58<18:57, 26.81it/s, loss=4.51, v_num=0]
Epoch 0:   9%|▉         | 3178/33666 [01:58<18:57, 26.81it/s, loss=4.51, v_num=0]
Epoch 0:   9%|▉         | 3178/33666 [01:58<18:57, 26.81it/s, loss=4.5, v_num=0]
Epoch 0:   9%|▉         | 3179/33666 [01:58<18:57, 26.81it/s, loss=4.5, v_num=0]
Epoch 0:   9%|▉         | 3179/33666 [01:58<18:57, 26.81it/s, loss=4.51, v_num=0]
Epoch 0:   9%|▉         | 3180/33666 [01:58<18:57, 26.81it/s, loss=4.51, v_num=0]
Epoch 0:   9%|▉         | 3180/33666 [01:58<18:57, 26.81it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 3181/33666 [01:58<18:57, 26.81it/s, loss=4.53, v_num=0]
Epoch 0:   9%|▉         | 3181/33666 [01:58<18:57, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3182/33666 [01:58<18:57, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3182/33666 [01:58<18:57, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3183/33666 [01:58<18:56, 26.81it/s, loss=4.55, v_num=0]
Epoch 0:   9%|▉         | 3183/33666 [01:58<18:56, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3184/33666 [01:58<18:56, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3184/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3185/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3185/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3186/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3186/33666 [01:58<18:56, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3187/33666 [01:58<18:56, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:   9%|▉         | 3187/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3188/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3188/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3189/33666 [01:58<18:56, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3189/33666 [01:58<18:56, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3190/33666 [01:58<18:56, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3190/33666 [01:58<18:56, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3191/33666 [01:59<18:56, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3191/33666 [01:59<18:56, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3192/33666 [01:59<18:56, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3192/33666 [01:59<18:56, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3193/33666 [01:59<18:56, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:   9%|▉         | 3193/33666 [01:59<18:56, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3194/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3194/33666 [01:59<18:56, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3195/33666 [01:59<18:56, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:   9%|▉         | 3195/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3196/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3196/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3197/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:   9%|▉         | 3197/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3198/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:   9%|▉         | 3198/33666 [01:59<18:56, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3199/33666 [01:59<18:56, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3199/33666 [01:59<18:56, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3200/33666 [01:59<18:56, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3200/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3201/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3201/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3202/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3202/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3203/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3203/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3204/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3204/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3205/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3205/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3206/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3206/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3207/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3207/33666 [01:59<18:56, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3208/33666 [01:59<18:56, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3208/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3209/33666 [01:59<18:56, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3209/33666 [01:59<18:56, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3210/33666 [01:59<18:56, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3210/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3211/33666 [01:59<18:56, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3211/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3212/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3212/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3213/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3213/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3214/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3214/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3215/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3215/33666 [01:59<18:56, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3216/33666 [01:59<18:56, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3216/33666 [01:59<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3217/33666 [02:00<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3217/33666 [02:00<18:56, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3218/33666 [02:00<18:56, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3218/33666 [02:00<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3219/33666 [02:00<18:56, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3219/33666 [02:00<18:56, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3220/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3220/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3221/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3221/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3222/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3222/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3223/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3223/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3224/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3224/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3225/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3225/33666 [02:00<18:55, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3226/33666 [02:00<18:55, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3226/33666 [02:00<18:55, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3227/33666 [02:00<18:55, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3227/33666 [02:00<18:55, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3228/33666 [02:00<18:55, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3228/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3229/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3229/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3230/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3230/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3231/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3231/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3232/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3232/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3233/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3233/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3234/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3234/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3235/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3235/33666 [02:00<18:55, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3236/33666 [02:00<18:55, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3236/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3237/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3237/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3238/33666 [02:00<18:55, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3238/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3239/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3239/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3240/33666 [02:00<18:55, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3240/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3241/33666 [02:00<18:55, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3241/33666 [02:00<18:55, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3242/33666 [02:00<18:55, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3242/33666 [02:00<18:55, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3243/33666 [02:00<18:55, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3243/33666 [02:00<18:55, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3244/33666 [02:01<18:55, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3244/33666 [02:01<18:55, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3245/33666 [02:01<18:54, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3245/33666 [02:01<18:54, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  10%|▉         | 3246/33666 [02:01<18:54, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  10%|▉         | 3246/33666 [02:01<18:54, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  10%|▉         | 3247/33666 [02:01<18:54, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  10%|▉         | 3247/33666 [02:01<18:54, 26.80it/s, loss=4.52, v_num=0]
Epoch 0:  10%|▉         | 3248/33666 [02:01<18:54, 26.80it/s, loss=4.52, v_num=0]
Epoch 0:  10%|▉         | 3248/33666 [02:01<18:54, 26.80it/s, loss=4.52, v_num=0]
Epoch 0:  10%|▉         | 3249/33666 [02:01<18:54, 26.80it/s, loss=4.52, v_num=0]
Epoch 0:  10%|▉         | 3249/33666 [02:01<18:54, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  10%|▉         | 3250/33666 [02:01<18:54, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  10%|▉         | 3250/33666 [02:01<18:54, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3251/33666 [02:01<18:54, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3251/33666 [02:01<18:54, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3252/33666 [02:01<18:54, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3252/33666 [02:01<18:54, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3253/33666 [02:01<18:54, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3253/33666 [02:01<18:54, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3254/33666 [02:01<18:54, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3254/33666 [02:01<18:54, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3255/33666 [02:01<18:54, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3255/33666 [02:01<18:54, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3256/33666 [02:01<18:54, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3256/33666 [02:01<18:54, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3257/33666 [02:01<18:54, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3257/33666 [02:01<18:54, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3258/33666 [02:01<18:54, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3258/33666 [02:01<18:54, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3259/33666 [02:01<18:54, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3259/33666 [02:01<18:54, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3260/33666 [02:01<18:54, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3260/33666 [02:01<18:54, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3261/33666 [02:01<18:54, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3261/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3262/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3262/33666 [02:01<18:54, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3263/33666 [02:01<18:54, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3263/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3264/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3264/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3265/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3265/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3266/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3266/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3267/33666 [02:01<18:54, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3267/33666 [02:01<18:54, 26.81it/s, loss=4.71, v_num=0]
Epoch 0:  10%|▉         | 3268/33666 [02:01<18:54, 26.81it/s, loss=4.71, v_num=0]
Epoch 0:  10%|▉         | 3268/33666 [02:01<18:54, 26.81it/s, loss=4.73, v_num=0]
Epoch 0:  10%|▉         | 3269/33666 [02:01<18:53, 26.81it/s, loss=4.73, v_num=0]
Epoch 0:  10%|▉         | 3269/33666 [02:01<18:53, 26.81it/s, loss=4.73, v_num=0]
Epoch 0:  10%|▉         | 3270/33666 [02:01<18:53, 26.81it/s, loss=4.73, v_num=0]
Epoch 0:  10%|▉         | 3270/33666 [02:01<18:53, 26.81it/s, loss=4.73, v_num=0]
Epoch 0:  10%|▉         | 3271/33666 [02:02<18:53, 26.81it/s, loss=4.73, v_num=0]
Epoch 0:  10%|▉         | 3271/33666 [02:02<18:53, 26.81it/s, loss=4.7, v_num=0]
Epoch 0:  10%|▉         | 3272/33666 [02:02<18:53, 26.81it/s, loss=4.7, v_num=0]
Epoch 0:  10%|▉         | 3272/33666 [02:02<18:53, 26.81it/s, loss=4.7, v_num=0]
Epoch 0:  10%|▉         | 3273/33666 [02:02<18:53, 26.81it/s, loss=4.7, v_num=0]
Epoch 0:  10%|▉         | 3273/33666 [02:02<18:53, 26.81it/s, loss=4.71, v_num=0]
Epoch 0:  10%|▉         | 3274/33666 [02:02<18:53, 26.81it/s, loss=4.71, v_num=0]
Epoch 0:  10%|▉         | 3274/33666 [02:02<18:53, 26.81it/s, loss=4.69, v_num=0]
Epoch 0:  10%|▉         | 3275/33666 [02:02<18:53, 26.81it/s, loss=4.69, v_num=0]
Epoch 0:  10%|▉         | 3275/33666 [02:02<18:53, 26.81it/s, loss=4.69, v_num=0]
Epoch 0:  10%|▉         | 3276/33666 [02:02<18:53, 26.81it/s, loss=4.69, v_num=0]
Epoch 0:  10%|▉         | 3276/33666 [02:02<18:53, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3277/33666 [02:02<18:53, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3277/33666 [02:02<18:53, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3278/33666 [02:02<18:53, 26.81it/s, loss=4.68, v_num=0]
Epoch 0:  10%|▉         | 3278/33666 [02:02<18:53, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3279/33666 [02:02<18:53, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3279/33666 [02:02<18:53, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3280/33666 [02:02<18:53, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3280/33666 [02:02<18:53, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3281/33666 [02:02<18:53, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3281/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3282/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3282/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3283/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3283/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3284/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3284/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3285/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3285/33666 [02:02<18:53, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3286/33666 [02:02<18:53, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3286/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3287/33666 [02:02<18:53, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3287/33666 [02:02<18:53, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3288/33666 [02:02<18:53, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3288/33666 [02:02<18:53, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3289/33666 [02:02<18:53, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3289/33666 [02:02<18:53, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3290/33666 [02:02<18:53, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3290/33666 [02:02<18:53, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3291/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3291/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3292/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3292/33666 [02:02<18:52, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3293/33666 [02:02<18:52, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3293/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3294/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3294/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3295/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3295/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3296/33666 [02:02<18:52, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3296/33666 [02:02<18:52, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3297/33666 [02:02<18:52, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3297/33666 [02:02<18:52, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3298/33666 [02:03<18:52, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3298/33666 [02:03<18:52, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3299/33666 [02:03<18:52, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3299/33666 [02:03<18:52, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3300/33666 [02:03<18:52, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3300/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3301/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3301/33666 [02:03<18:52, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3302/33666 [02:03<18:52, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3302/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3303/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3303/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3304/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3304/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3305/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3305/33666 [02:03<18:52, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3306/33666 [02:03<18:52, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3306/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3307/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3307/33666 [02:03<18:52, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3308/33666 [02:03<18:52, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3308/33666 [02:03<18:52, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3309/33666 [02:03<18:52, 26.81it/s, loss=4.67, v_num=0]
Epoch 0:  10%|▉         | 3309/33666 [02:03<18:52, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3310/33666 [02:03<18:52, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3310/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3311/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3311/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3312/33666 [02:03<18:52, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3312/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3313/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3313/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3314/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3314/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3315/33666 [02:03<18:52, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3315/33666 [02:03<18:52, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3316/33666 [02:03<18:52, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3316/33666 [02:03<18:52, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3317/33666 [02:03<18:51, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3317/33666 [02:03<18:51, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3318/33666 [02:03<18:51, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3318/33666 [02:03<18:51, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3319/33666 [02:03<18:51, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3319/33666 [02:03<18:51, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3320/33666 [02:03<18:51, 26.81it/s, loss=4.65, v_num=0]
Epoch 0:  10%|▉         | 3320/33666 [02:03<18:51, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3321/33666 [02:03<18:51, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3321/33666 [02:03<18:51, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3322/33666 [02:03<18:51, 26.81it/s, loss=4.66, v_num=0]
Epoch 0:  10%|▉         | 3322/33666 [02:03<18:51, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3323/33666 [02:03<18:51, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3323/33666 [02:03<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3324/33666 [02:03<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3324/33666 [02:03<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3325/33666 [02:04<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3325/33666 [02:04<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3326/33666 [02:04<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3326/33666 [02:04<18:51, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3327/33666 [02:04<18:51, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3327/33666 [02:04<18:51, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3328/33666 [02:04<18:51, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3328/33666 [02:04<18:51, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3329/33666 [02:04<18:51, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3329/33666 [02:04<18:51, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3330/33666 [02:04<18:51, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3330/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3331/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3331/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3332/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3332/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3333/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3333/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3334/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3334/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3335/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3335/33666 [02:04<18:51, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3336/33666 [02:04<18:51, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3336/33666 [02:04<18:51, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3337/33666 [02:04<18:51, 26.81it/s, loss=4.64, v_num=0]
Epoch 0:  10%|▉         | 3337/33666 [02:04<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3338/33666 [02:04<18:51, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  10%|▉         | 3338/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3339/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3339/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3340/33666 [02:04<18:51, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3340/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3341/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3341/33666 [02:04<18:51, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3342/33666 [02:04<18:50, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3342/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3343/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3343/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3344/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3344/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3345/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3345/33666 [02:04<18:50, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3346/33666 [02:04<18:50, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3346/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3347/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3347/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3348/33666 [02:04<18:50, 26.81it/s, loss=4.62, v_num=0]
Epoch 0:  10%|▉         | 3348/33666 [02:04<18:50, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3349/33666 [02:04<18:50, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3349/33666 [02:04<18:50, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3350/33666 [02:04<18:50, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|▉         | 3350/33666 [02:04<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3351/33666 [02:04<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3351/33666 [02:04<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3352/33666 [02:05<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3352/33666 [02:05<18:50, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3353/33666 [02:05<18:50, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3353/33666 [02:05<18:50, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3354/33666 [02:05<18:50, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3354/33666 [02:05<18:50, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3355/33666 [02:05<18:50, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3355/33666 [02:05<18:50, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3356/33666 [02:05<18:50, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3356/33666 [02:05<18:50, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3357/33666 [02:05<18:50, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  10%|▉         | 3357/33666 [02:05<18:50, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3358/33666 [02:05<18:50, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3358/33666 [02:05<18:50, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3359/33666 [02:05<18:50, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3359/33666 [02:05<18:50, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3360/33666 [02:05<18:50, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|▉         | 3360/33666 [02:05<18:50, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|▉         | 3361/33666 [02:05<18:50, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|▉         | 3361/33666 [02:05<18:50, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3362/33666 [02:05<18:50, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3362/33666 [02:05<18:50, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3363/33666 [02:05<18:50, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  10%|▉         | 3363/33666 [02:05<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3364/33666 [02:05<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3364/33666 [02:05<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3365/33666 [02:05<18:50, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  10%|▉         | 3365/33666 [02:05<18:50, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3366/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|▉         | 3366/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3367/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3367/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3368/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3368/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3369/33666 [02:05<18:49, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3369/33666 [02:05<18:49, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3370/33666 [02:05<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3370/33666 [02:05<18:49, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3371/33666 [02:05<18:49, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3371/33666 [02:05<18:49, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3372/33666 [02:05<18:49, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3372/33666 [02:05<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3373/33666 [02:05<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3373/33666 [02:05<18:49, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3374/33666 [02:05<18:49, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3374/33666 [02:05<18:49, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3375/33666 [02:05<18:49, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3375/33666 [02:05<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3376/33666 [02:05<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3376/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3377/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3377/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3378/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3378/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3379/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3379/33666 [02:05<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3380/33666 [02:06<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3380/33666 [02:06<18:49, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3381/33666 [02:06<18:49, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3381/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3382/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3382/33666 [02:06<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3383/33666 [02:06<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3383/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3384/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3384/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3385/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3385/33666 [02:06<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3386/33666 [02:06<18:49, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3386/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3387/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3387/33666 [02:06<18:49, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3388/33666 [02:06<18:49, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3388/33666 [02:06<18:49, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3389/33666 [02:06<18:48, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3389/33666 [02:06<18:48, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3390/33666 [02:06<18:48, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3390/33666 [02:06<18:48, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3391/33666 [02:06<18:48, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3391/33666 [02:06<18:48, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3392/33666 [02:06<18:48, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3392/33666 [02:06<18:48, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3393/33666 [02:06<18:48, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3393/33666 [02:06<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3394/33666 [02:06<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3394/33666 [02:06<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3395/33666 [02:06<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3395/33666 [02:06<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3396/33666 [02:06<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3396/33666 [02:06<18:48, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  10%|█         | 3397/33666 [02:06<18:48, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  10%|█         | 3397/33666 [02:06<18:48, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  10%|█         | 3398/33666 [02:06<18:48, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  10%|█         | 3398/33666 [02:06<18:48, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  10%|█         | 3399/33666 [02:06<18:48, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  10%|█         | 3399/33666 [02:06<18:48, 26.82it/s, loss=4.65, v_num=0]
Epoch 0:  10%|█         | 3400/33666 [02:06<18:48, 26.82it/s, loss=4.65, v_num=0]
Epoch 0:  10%|█         | 3400/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3401/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3401/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3402/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3402/33666 [02:06<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3403/33666 [02:06<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3403/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3404/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3404/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3405/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3405/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3406/33666 [02:06<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3406/33666 [02:06<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3407/33666 [02:07<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3407/33666 [02:07<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3408/33666 [02:07<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3408/33666 [02:07<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3409/33666 [02:07<18:48, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3409/33666 [02:07<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3410/33666 [02:07<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3410/33666 [02:07<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3411/33666 [02:07<18:48, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3411/33666 [02:07<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3412/33666 [02:07<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3412/33666 [02:07<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3413/33666 [02:07<18:48, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3413/33666 [02:07<18:48, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3414/33666 [02:07<18:47, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3414/33666 [02:07<18:47, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3415/33666 [02:07<18:47, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3415/33666 [02:07<18:47, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3416/33666 [02:07<18:47, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3416/33666 [02:07<18:47, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3417/33666 [02:07<18:47, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3417/33666 [02:07<18:47, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3418/33666 [02:07<18:47, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3418/33666 [02:07<18:47, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3419/33666 [02:07<18:47, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3419/33666 [02:07<18:47, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3420/33666 [02:07<18:47, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3420/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3421/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3421/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3422/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3422/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3423/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3423/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3424/33666 [02:07<18:47, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3424/33666 [02:07<18:47, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3425/33666 [02:07<18:47, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3425/33666 [02:07<18:47, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3426/33666 [02:07<18:47, 26.82it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3426/33666 [02:07<18:47, 26.82it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3427/33666 [02:07<18:47, 26.82it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3427/33666 [02:07<18:47, 26.82it/s, loss=4.53, v_num=0]
Epoch 0:  10%|█         | 3428/33666 [02:07<18:47, 26.82it/s, loss=4.53, v_num=0]
Epoch 0:  10%|█         | 3428/33666 [02:07<18:47, 26.82it/s, loss=4.53, v_num=0]
Epoch 0:  10%|█         | 3429/33666 [02:07<18:47, 26.82it/s, loss=4.53, v_num=0]
Epoch 0:  10%|█         | 3429/33666 [02:07<18:47, 26.82it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3430/33666 [02:07<18:47, 26.82it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3430/33666 [02:07<18:47, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3431/33666 [02:07<18:47, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3431/33666 [02:07<18:47, 26.82it/s, loss=4.53, v_num=0]
Epoch 0:  10%|█         | 3432/33666 [02:07<18:47, 26.82it/s, loss=4.53, v_num=0]
Epoch 0:  10%|█         | 3432/33666 [02:07<18:47, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3433/33666 [02:07<18:47, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3433/33666 [02:07<18:47, 26.82it/s, loss=4.5, v_num=0]
Epoch 0:  10%|█         | 3434/33666 [02:08<18:47, 26.82it/s, loss=4.5, v_num=0]
Epoch 0:  10%|█         | 3434/33666 [02:08<18:47, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3435/33666 [02:08<18:46, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3435/33666 [02:08<18:46, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3436/33666 [02:08<18:46, 26.82it/s, loss=4.52, v_num=0]
Epoch 0:  10%|█         | 3436/33666 [02:08<18:46, 26.82it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3437/33666 [02:08<18:46, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3437/33666 [02:08<18:46, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3438/33666 [02:08<18:46, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3438/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3439/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3439/33666 [02:08<18:46, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3440/33666 [02:08<18:46, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3440/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3441/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3441/33666 [02:08<18:46, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3442/33666 [02:08<18:46, 26.82it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3442/33666 [02:08<18:46, 26.82it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3443/33666 [02:08<18:46, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3443/33666 [02:08<18:46, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3444/33666 [02:08<18:46, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3444/33666 [02:08<18:46, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3445/33666 [02:08<18:46, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3445/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3446/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3446/33666 [02:08<18:46, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3447/33666 [02:08<18:46, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3447/33666 [02:08<18:46, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3448/33666 [02:08<18:46, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3448/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3449/33666 [02:08<18:46, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3449/33666 [02:08<18:46, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3450/33666 [02:08<18:46, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3450/33666 [02:08<18:46, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3451/33666 [02:08<18:46, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3451/33666 [02:08<18:46, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3452/33666 [02:08<18:46, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3452/33666 [02:08<18:46, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3453/33666 [02:08<18:46, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3453/33666 [02:08<18:46, 26.83it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3454/33666 [02:08<18:46, 26.83it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3454/33666 [02:08<18:46, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3455/33666 [02:08<18:46, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3455/33666 [02:08<18:46, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3456/33666 [02:08<18:46, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3456/33666 [02:08<18:46, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3457/33666 [02:08<18:46, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3457/33666 [02:08<18:46, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3458/33666 [02:08<18:45, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3458/33666 [02:08<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3459/33666 [02:08<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3459/33666 [02:08<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3460/33666 [02:08<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3460/33666 [02:08<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3461/33666 [02:09<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3461/33666 [02:09<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3462/33666 [02:09<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3462/33666 [02:09<18:45, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3463/33666 [02:09<18:45, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3463/33666 [02:09<18:45, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3464/33666 [02:09<18:45, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3464/33666 [02:09<18:45, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3465/33666 [02:09<18:45, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3465/33666 [02:09<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3466/33666 [02:09<18:45, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  10%|█         | 3466/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3467/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3467/33666 [02:09<18:45, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3468/33666 [02:09<18:45, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3468/33666 [02:09<18:45, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3469/33666 [02:09<18:45, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3469/33666 [02:09<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3470/33666 [02:09<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3470/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3471/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3471/33666 [02:09<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3472/33666 [02:09<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3472/33666 [02:09<18:45, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3473/33666 [02:09<18:45, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3473/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3474/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3474/33666 [02:09<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3475/33666 [02:09<18:45, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3475/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3476/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3476/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3477/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3477/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3478/33666 [02:09<18:45, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3478/33666 [02:09<18:45, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3479/33666 [02:09<18:45, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3479/33666 [02:09<18:45, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3480/33666 [02:09<18:45, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3480/33666 [02:09<18:45, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3481/33666 [02:09<18:45, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3481/33666 [02:09<18:45, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3482/33666 [02:09<18:44, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3482/33666 [02:09<18:44, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3483/33666 [02:09<18:44, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3483/33666 [02:09<18:44, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3484/33666 [02:09<18:44, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3484/33666 [02:09<18:44, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3485/33666 [02:09<18:44, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3485/33666 [02:09<18:44, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3486/33666 [02:09<18:44, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3486/33666 [02:09<18:44, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3487/33666 [02:09<18:44, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3487/33666 [02:09<18:44, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3488/33666 [02:09<18:44, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3488/33666 [02:09<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3489/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3489/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3490/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3490/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3491/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3491/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3492/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3492/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3493/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3493/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3494/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3494/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3495/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3495/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3496/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3496/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3497/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3497/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3498/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3498/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3499/33666 [02:10<18:44, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  10%|█         | 3499/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3500/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3500/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3501/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3501/33666 [02:10<18:44, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3502/33666 [02:10<18:44, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  10%|█         | 3502/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3503/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3503/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3504/33666 [02:10<18:44, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  10%|█         | 3504/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3505/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3505/33666 [02:10<18:44, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3506/33666 [02:10<18:44, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3506/33666 [02:10<18:44, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3507/33666 [02:10<18:43, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3507/33666 [02:10<18:43, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3508/33666 [02:10<18:43, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3508/33666 [02:10<18:44, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3509/33666 [02:10<18:43, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3509/33666 [02:10<18:43, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3510/33666 [02:10<18:43, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3510/33666 [02:10<18:43, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3511/33666 [02:10<18:43, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3511/33666 [02:10<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3512/33666 [02:10<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3512/33666 [02:10<18:43, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3513/33666 [02:10<18:43, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3513/33666 [02:10<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3514/33666 [02:10<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3514/33666 [02:10<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3515/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3515/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3516/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3516/33666 [02:11<18:43, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3517/33666 [02:11<18:43, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3517/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3518/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3518/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3519/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3519/33666 [02:11<18:43, 26.83it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3520/33666 [02:11<18:43, 26.83it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3520/33666 [02:11<18:43, 26.83it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3521/33666 [02:11<18:43, 26.83it/s, loss=4.64, v_num=0]
Epoch 0:  10%|█         | 3521/33666 [02:11<18:43, 26.83it/s, loss=4.65, v_num=0]
Epoch 0:  10%|█         | 3522/33666 [02:11<18:43, 26.83it/s, loss=4.65, v_num=0]
Epoch 0:  10%|█         | 3522/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3523/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3523/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3524/33666 [02:11<18:43, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  10%|█         | 3524/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3525/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3525/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3526/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3526/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3527/33666 [02:11<18:43, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  10%|█         | 3527/33666 [02:11<18:43, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3528/33666 [02:11<18:43, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3528/33666 [02:11<18:43, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3529/33666 [02:11<18:43, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3529/33666 [02:11<18:43, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3530/33666 [02:11<18:43, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  10%|█         | 3530/33666 [02:11<18:43, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3531/33666 [02:11<18:43, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  10%|█         | 3531/33666 [02:11<18:43, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3532/33666 [02:11<18:43, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3532/33666 [02:11<18:43, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3533/33666 [02:11<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  10%|█         | 3533/33666 [02:11<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3534/33666 [02:11<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  10%|█         | 3534/33666 [02:11<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3535/33666 [02:11<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3535/33666 [02:11<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3536/33666 [02:11<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3536/33666 [02:11<18:42, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3537/33666 [02:11<18:42, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3537/33666 [02:11<18:42, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3538/33666 [02:11<18:42, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3538/33666 [02:11<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3539/33666 [02:11<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3539/33666 [02:11<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3540/33666 [02:11<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3540/33666 [02:11<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3541/33666 [02:11<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3541/33666 [02:11<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3542/33666 [02:12<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3542/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3543/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3543/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3544/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3544/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3545/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3545/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3546/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3546/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3547/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3547/33666 [02:12<18:42, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3548/33666 [02:12<18:42, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3548/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3549/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3549/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3550/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3550/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3551/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3551/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3552/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3552/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3553/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3553/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3554/33666 [02:12<18:42, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3554/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3555/33666 [02:12<18:42, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3555/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3556/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3556/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3557/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3557/33666 [02:12<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3558/33666 [02:12<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3558/33666 [02:12<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3559/33666 [02:12<18:42, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3559/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3560/33666 [02:12<18:42, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3560/33666 [02:12<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3561/33666 [02:12<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3561/33666 [02:12<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3562/33666 [02:12<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3562/33666 [02:12<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3563/33666 [02:12<18:42, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3563/33666 [02:12<18:42, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3564/33666 [02:12<18:42, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3564/33666 [02:12<18:42, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3565/33666 [02:12<18:42, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3565/33666 [02:12<18:42, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3566/33666 [02:12<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3566/33666 [02:12<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3567/33666 [02:12<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3567/33666 [02:12<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3568/33666 [02:12<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3568/33666 [02:12<18:41, 26.83it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3569/33666 [02:13<18:41, 26.83it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3569/33666 [02:13<18:41, 26.83it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3570/33666 [02:13<18:41, 26.83it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3570/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3571/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3571/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3572/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3572/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3573/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3573/33666 [02:13<18:41, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3574/33666 [02:13<18:41, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3574/33666 [02:13<18:41, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3575/33666 [02:13<18:41, 26.83it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3575/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3576/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3576/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3577/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3577/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3578/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3578/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3579/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3579/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3580/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3580/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3581/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3581/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3582/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3582/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3583/33666 [02:13<18:41, 26.83it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3583/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3584/33666 [02:13<18:41, 26.83it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3584/33666 [02:13<18:41, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3585/33666 [02:13<18:41, 26.83it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3585/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3586/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3586/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3587/33666 [02:13<18:41, 26.83it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3587/33666 [02:13<18:41, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3588/33666 [02:13<18:41, 26.83it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3588/33666 [02:13<18:41, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3589/33666 [02:13<18:41, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3589/33666 [02:13<18:41, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3590/33666 [02:13<18:41, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3590/33666 [02:13<18:41, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3591/33666 [02:13<18:40, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3591/33666 [02:13<18:40, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3592/33666 [02:13<18:40, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3592/33666 [02:13<18:40, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3593/33666 [02:13<18:40, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3593/33666 [02:13<18:40, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3594/33666 [02:13<18:40, 26.83it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3594/33666 [02:13<18:40, 26.83it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3595/33666 [02:13<18:40, 26.83it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3595/33666 [02:13<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3596/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3596/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3597/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3597/33666 [02:14<18:40, 26.83it/s, loss=4.69, v_num=0]
Epoch 0:  11%|█         | 3598/33666 [02:14<18:40, 26.83it/s, loss=4.69, v_num=0]
Epoch 0:  11%|█         | 3598/33666 [02:14<18:40, 26.83it/s, loss=4.71, v_num=0]
Epoch 0:  11%|█         | 3599/33666 [02:14<18:40, 26.83it/s, loss=4.71, v_num=0]
Epoch 0:  11%|█         | 3599/33666 [02:14<18:40, 26.83it/s, loss=4.7, v_num=0]
Epoch 0:  11%|█         | 3600/33666 [02:14<18:40, 26.83it/s, loss=4.7, v_num=0]
Epoch 0:  11%|█         | 3600/33666 [02:14<18:40, 26.83it/s, loss=4.71, v_num=0]
Epoch 0:  11%|█         | 3601/33666 [02:14<18:40, 26.83it/s, loss=4.71, v_num=0]
Epoch 0:  11%|█         | 3601/33666 [02:14<18:40, 26.83it/s, loss=4.73, v_num=0]
Epoch 0:  11%|█         | 3602/33666 [02:14<18:40, 26.83it/s, loss=4.73, v_num=0]
Epoch 0:  11%|█         | 3602/33666 [02:14<18:40, 26.83it/s, loss=4.72, v_num=0]
Epoch 0:  11%|█         | 3603/33666 [02:14<18:40, 26.83it/s, loss=4.72, v_num=0]
Epoch 0:  11%|█         | 3603/33666 [02:14<18:40, 26.83it/s, loss=4.71, v_num=0]
Epoch 0:  11%|█         | 3604/33666 [02:14<18:40, 26.83it/s, loss=4.71, v_num=0]
Epoch 0:  11%|█         | 3604/33666 [02:14<18:40, 26.83it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3605/33666 [02:14<18:40, 26.83it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3605/33666 [02:14<18:40, 26.83it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3606/33666 [02:14<18:40, 26.83it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3606/33666 [02:14<18:40, 26.83it/s, loss=4.69, v_num=0]
Epoch 0:  11%|█         | 3607/33666 [02:14<18:40, 26.83it/s, loss=4.69, v_num=0]
Epoch 0:  11%|█         | 3607/33666 [02:14<18:40, 26.83it/s, loss=4.69, v_num=0]
Epoch 0:  11%|█         | 3608/33666 [02:14<18:40, 26.83it/s, loss=4.69, v_num=0]
Epoch 0:  11%|█         | 3608/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3609/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3609/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3610/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3610/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3611/33666 [02:14<18:40, 26.83it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3611/33666 [02:14<18:40, 26.83it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3612/33666 [02:14<18:40, 26.83it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3612/33666 [02:14<18:40, 26.83it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3613/33666 [02:14<18:40, 26.83it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3613/33666 [02:14<18:40, 26.83it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3614/33666 [02:14<18:40, 26.83it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3614/33666 [02:14<18:40, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3615/33666 [02:14<18:40, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3615/33666 [02:14<18:40, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3616/33666 [02:14<18:40, 26.83it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3616/33666 [02:14<18:40, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3617/33666 [02:14<18:40, 26.83it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3617/33666 [02:14<18:40, 26.83it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3618/33666 [02:14<18:40, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3618/33666 [02:14<18:40, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3619/33666 [02:14<18:40, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3619/33666 [02:14<18:40, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3620/33666 [02:14<18:40, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3620/33666 [02:14<18:40, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3621/33666 [02:14<18:40, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3621/33666 [02:14<18:40, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3622/33666 [02:15<18:40, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3622/33666 [02:15<18:40, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3623/33666 [02:15<18:39, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3623/33666 [02:15<18:39, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3624/33666 [02:15<18:39, 26.82it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3624/33666 [02:15<18:39, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3625/33666 [02:15<18:39, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3625/33666 [02:15<18:39, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3626/33666 [02:15<18:39, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3626/33666 [02:15<18:39, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3627/33666 [02:15<18:39, 26.82it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3627/33666 [02:15<18:39, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3628/33666 [02:15<18:39, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3628/33666 [02:15<18:39, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3629/33666 [02:15<18:39, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3629/33666 [02:15<18:39, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3630/33666 [02:15<18:39, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3630/33666 [02:15<18:39, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3631/33666 [02:15<18:39, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3631/33666 [02:15<18:39, 26.82it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3632/33666 [02:15<18:39, 26.82it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3632/33666 [02:15<18:39, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3633/33666 [02:15<18:39, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3633/33666 [02:15<18:39, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3634/33666 [02:15<18:39, 26.82it/s, loss=4.66, v_num=0]
Epoch 0:  11%|█         | 3634/33666 [02:15<18:39, 26.82it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3635/33666 [02:15<18:39, 26.82it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3635/33666 [02:15<18:39, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3636/33666 [02:15<18:39, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3636/33666 [02:15<18:39, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3637/33666 [02:15<18:39, 26.82it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3637/33666 [02:15<18:39, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3638/33666 [02:15<18:39, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3638/33666 [02:15<18:39, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3639/33666 [02:15<18:39, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3639/33666 [02:15<18:39, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3640/33666 [02:15<18:39, 26.82it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3640/33666 [02:15<18:39, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3641/33666 [02:15<18:39, 26.82it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3641/33666 [02:15<18:39, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3642/33666 [02:15<18:39, 26.82it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3642/33666 [02:15<18:39, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3643/33666 [02:15<18:39, 26.82it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3643/33666 [02:15<18:39, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3644/33666 [02:15<18:39, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3644/33666 [02:15<18:39, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3645/33666 [02:15<18:39, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3645/33666 [02:15<18:39, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3646/33666 [02:15<18:39, 26.81it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3646/33666 [02:15<18:39, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3647/33666 [02:16<18:39, 26.81it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3647/33666 [02:16<18:39, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3648/33666 [02:16<18:39, 26.81it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3648/33666 [02:16<18:39, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3649/33666 [02:16<18:39, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3649/33666 [02:16<18:39, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3650/33666 [02:16<18:39, 26.81it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3650/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3651/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3651/33666 [02:16<18:39, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3652/33666 [02:16<18:39, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3652/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3653/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3653/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3654/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3654/33666 [02:16<18:39, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3655/33666 [02:16<18:39, 26.81it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3655/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3656/33666 [02:16<18:39, 26.81it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3656/33666 [02:16<18:39, 26.81it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3657/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3657/33666 [02:16<18:39, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3658/33666 [02:16<18:39, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3658/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3659/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3659/33666 [02:16<18:39, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3660/33666 [02:16<18:39, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3660/33666 [02:16<18:39, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3661/33666 [02:16<18:39, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3661/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3662/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3662/33666 [02:16<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3663/33666 [02:16<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3663/33666 [02:16<18:39, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3664/33666 [02:16<18:39, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3664/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3665/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3665/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3666/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3666/33666 [02:16<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3667/33666 [02:16<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3667/33666 [02:16<18:39, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3668/33666 [02:16<18:39, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3668/33666 [02:16<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3669/33666 [02:16<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3669/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3670/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3670/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3671/33666 [02:16<18:39, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3671/33666 [02:16<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3672/33666 [02:17<18:39, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3672/33666 [02:17<18:39, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3673/33666 [02:17<18:39, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3673/33666 [02:17<18:39, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3674/33666 [02:17<18:39, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3674/33666 [02:17<18:39, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3675/33666 [02:17<18:39, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3675/33666 [02:17<18:39, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3676/33666 [02:17<18:39, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3676/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3677/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3677/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3678/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3678/33666 [02:17<18:39, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3679/33666 [02:17<18:39, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3679/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3680/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3680/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3681/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3681/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3682/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3682/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3683/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3683/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3684/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3684/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3685/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3685/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3686/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3686/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3687/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3687/33666 [02:17<18:39, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3688/33666 [02:17<18:39, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3688/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3689/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3689/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3690/33666 [02:17<18:39, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3690/33666 [02:17<18:39, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3691/33666 [02:17<18:39, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3691/33666 [02:17<18:39, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3692/33666 [02:17<18:39, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3692/33666 [02:17<18:39, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3693/33666 [02:17<18:38, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3693/33666 [02:17<18:38, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3694/33666 [02:17<18:38, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3694/33666 [02:17<18:38, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3695/33666 [02:17<18:38, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3695/33666 [02:17<18:38, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3696/33666 [02:17<18:38, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3696/33666 [02:17<18:38, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3697/33666 [02:18<18:38, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3697/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3698/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3698/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3699/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3699/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3700/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3700/33666 [02:18<18:38, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3701/33666 [02:18<18:38, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3701/33666 [02:18<18:38, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3702/33666 [02:18<18:38, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3702/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3703/33666 [02:18<18:38, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3703/33666 [02:18<18:38, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3704/33666 [02:18<18:38, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3704/33666 [02:18<18:38, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3705/33666 [02:18<18:38, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3705/33666 [02:18<18:38, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3706/33666 [02:18<18:38, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3706/33666 [02:18<18:38, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3707/33666 [02:18<18:38, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3707/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3708/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3708/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3709/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3709/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3710/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3710/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3711/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3711/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3712/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3712/33666 [02:18<18:38, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3713/33666 [02:18<18:38, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3713/33666 [02:18<18:38, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3714/33666 [02:18<18:38, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3714/33666 [02:18<18:38, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3715/33666 [02:18<18:38, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3715/33666 [02:18<18:38, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3716/33666 [02:18<18:38, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3716/33666 [02:18<18:38, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3717/33666 [02:18<18:38, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3717/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3718/33666 [02:18<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3718/33666 [02:18<18:38, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3719/33666 [02:18<18:38, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3719/33666 [02:18<18:38, 26.78it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3720/33666 [02:18<18:38, 26.78it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3720/33666 [02:18<18:38, 26.78it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3721/33666 [02:18<18:38, 26.78it/s, loss=4.53, v_num=0]
Epoch 0:  11%|█         | 3721/33666 [02:18<18:38, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3722/33666 [02:19<18:38, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3722/33666 [02:19<18:38, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3723/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3723/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3724/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3724/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3725/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3725/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3726/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3726/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3727/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3727/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3728/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3728/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3729/33666 [02:19<18:38, 26.77it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3729/33666 [02:19<18:38, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3730/33666 [02:19<18:38, 26.77it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3730/33666 [02:19<18:38, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3731/33666 [02:19<18:38, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3731/33666 [02:19<18:38, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3732/33666 [02:19<18:38, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3732/33666 [02:19<18:38, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3733/33666 [02:19<18:38, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3733/33666 [02:19<18:38, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3734/33666 [02:19<18:38, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3734/33666 [02:19<18:38, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3735/33666 [02:19<18:38, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3735/33666 [02:19<18:38, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3736/33666 [02:19<18:38, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3736/33666 [02:19<18:38, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3737/33666 [02:19<18:38, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3737/33666 [02:19<18:38, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3738/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3738/33666 [02:19<18:37, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3739/33666 [02:19<18:37, 26.77it/s, loss=4.67, v_num=0]
Epoch 0:  11%|█         | 3739/33666 [02:19<18:37, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3740/33666 [02:19<18:37, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3740/33666 [02:19<18:37, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3741/33666 [02:19<18:37, 26.77it/s, loss=4.68, v_num=0]
Epoch 0:  11%|█         | 3741/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3742/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3742/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3743/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3743/33666 [02:19<18:37, 26.77it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3744/33666 [02:19<18:37, 26.77it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3744/33666 [02:19<18:37, 26.77it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3745/33666 [02:19<18:37, 26.77it/s, loss=4.65, v_num=0]
Epoch 0:  11%|█         | 3745/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3746/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3746/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3747/33666 [02:19<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3747/33666 [02:19<18:37, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3748/33666 [02:20<18:37, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3748/33666 [02:20<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3749/33666 [02:20<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3749/33666 [02:20<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3750/33666 [02:20<18:37, 26.77it/s, loss=4.64, v_num=0]
Epoch 0:  11%|█         | 3750/33666 [02:20<18:37, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3751/33666 [02:20<18:37, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3751/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3752/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3752/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3753/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3753/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3754/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3754/33666 [02:20<18:37, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3755/33666 [02:20<18:37, 26.77it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█         | 3755/33666 [02:20<18:37, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3756/33666 [02:20<18:37, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3756/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3757/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3757/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3758/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3758/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3759/33666 [02:20<18:37, 26.77it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█         | 3759/33666 [02:20<18:37, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3760/33666 [02:20<18:36, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3760/33666 [02:20<18:36, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3761/33666 [02:20<18:36, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3761/33666 [02:20<18:36, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3762/33666 [02:20<18:36, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3762/33666 [02:20<18:36, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3763/33666 [02:20<18:36, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3763/33666 [02:20<18:36, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3764/33666 [02:20<18:36, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3764/33666 [02:20<18:36, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3765/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3765/33666 [02:20<18:36, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3766/33666 [02:20<18:36, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3766/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3767/33666 [02:20<18:36, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3767/33666 [02:20<18:36, 26.77it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3768/33666 [02:20<18:36, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3768/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3769/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3769/33666 [02:20<18:36, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3770/33666 [02:20<18:36, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3770/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3771/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3771/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3772/33666 [02:20<18:36, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█         | 3772/33666 [02:20<18:36, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3773/33666 [02:20<18:36, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3773/33666 [02:20<18:36, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3774/33666 [02:20<18:36, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█         | 3774/33666 [02:20<18:36, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3775/33666 [02:20<18:36, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3775/33666 [02:20<18:36, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3776/33666 [02:21<18:36, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3776/33666 [02:21<18:36, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3777/33666 [02:21<18:36, 26.78it/s, loss=4.54, v_num=0]
Epoch 0:  11%|█         | 3777/33666 [02:21<18:36, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3778/33666 [02:21<18:36, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3778/33666 [02:21<18:36, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3779/33666 [02:21<18:36, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█         | 3779/33666 [02:21<18:36, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3780/33666 [02:21<18:36, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█         | 3780/33666 [02:21<18:36, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3781/33666 [02:21<18:36, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3781/33666 [02:21<18:36, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3782/33666 [02:21<18:36, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█         | 3782/33666 [02:21<18:36, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3783/33666 [02:21<18:36, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3783/33666 [02:21<18:36, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3784/33666 [02:21<18:35, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3784/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3785/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3785/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3786/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█         | 3786/33666 [02:21<18:35, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3787/33666 [02:21<18:35, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█         | 3787/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3788/33666 [02:21<18:35, 26.77it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3788/33666 [02:21<18:35, 26.77it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3789/33666 [02:21<18:35, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3789/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3790/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3790/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3791/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3791/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3792/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3792/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3793/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3793/33666 [02:21<18:35, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3794/33666 [02:21<18:35, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3794/33666 [02:21<18:35, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3795/33666 [02:21<18:35, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3795/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3796/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3796/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3797/33666 [02:21<18:35, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3797/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3798/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3798/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3799/33666 [02:21<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3799/33666 [02:21<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3800/33666 [02:21<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3800/33666 [02:21<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3801/33666 [02:21<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3801/33666 [02:21<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3802/33666 [02:21<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3802/33666 [02:21<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3803/33666 [02:22<18:35, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3803/33666 [02:22<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3804/33666 [02:22<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3804/33666 [02:22<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3805/33666 [02:22<18:35, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3805/33666 [02:22<18:35, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3806/33666 [02:22<18:34, 26.78it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3806/33666 [02:22<18:34, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3807/33666 [02:22<18:34, 26.78it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3807/33666 [02:22<18:34, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3808/33666 [02:22<18:34, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3808/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3809/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3809/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3810/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3810/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3811/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3811/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3812/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3812/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3813/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3813/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3814/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3814/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3815/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3815/33666 [02:22<18:34, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█▏        | 3816/33666 [02:22<18:34, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█▏        | 3816/33666 [02:22<18:34, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█▏        | 3817/33666 [02:22<18:34, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█▏        | 3817/33666 [02:22<18:34, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█▏        | 3818/33666 [02:22<18:34, 26.78it/s, loss=4.63, v_num=0]
Epoch 0:  11%|█▏        | 3818/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3819/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3819/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3820/33666 [02:22<18:34, 26.78it/s, loss=4.62, v_num=0]
Epoch 0:  11%|█▏        | 3820/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3821/33666 [02:22<18:34, 26.78it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3821/33666 [02:22<18:34, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█▏        | 3822/33666 [02:22<18:34, 26.78it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█▏        | 3822/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3823/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3823/33666 [02:22<18:34, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3824/33666 [02:22<18:34, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3824/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3825/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3825/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3826/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3826/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3827/33666 [02:22<18:34, 26.78it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3827/33666 [02:22<18:34, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3828/33666 [02:22<18:34, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3828/33666 [02:22<18:34, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3829/33666 [02:22<18:33, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3829/33666 [02:22<18:33, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3830/33666 [02:22<18:33, 26.78it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3830/33666 [02:22<18:33, 26.78it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3831/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3831/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3832/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3832/33666 [02:23<18:33, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3833/33666 [02:23<18:33, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3833/33666 [02:23<18:33, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3834/33666 [02:23<18:33, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3834/33666 [02:23<18:33, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3835/33666 [02:23<18:33, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3835/33666 [02:23<18:33, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3836/33666 [02:23<18:33, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3836/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3837/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3837/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3838/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3838/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3839/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3839/33666 [02:23<18:33, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█▏        | 3840/33666 [02:23<18:33, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█▏        | 3840/33666 [02:23<18:33, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█▏        | 3841/33666 [02:23<18:33, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  11%|█▏        | 3841/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3842/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3842/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3843/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3843/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3844/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3844/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3845/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3845/33666 [02:23<18:33, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3846/33666 [02:23<18:33, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3846/33666 [02:23<18:33, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3847/33666 [02:23<18:33, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3847/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3848/33666 [02:23<18:33, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3848/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3849/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3849/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3850/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3850/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3851/33666 [02:23<18:33, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3851/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3852/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3852/33666 [02:23<18:33, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3853/33666 [02:23<18:32, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3853/33666 [02:23<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3854/33666 [02:23<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3854/33666 [02:23<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3855/33666 [02:23<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3855/33666 [02:23<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3856/33666 [02:23<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3856/33666 [02:23<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3857/33666 [02:23<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3857/33666 [02:23<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3858/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3858/33666 [02:24<18:32, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3859/33666 [02:24<18:32, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3859/33666 [02:24<18:32, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3860/33666 [02:24<18:32, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  11%|█▏        | 3860/33666 [02:24<18:32, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3861/33666 [02:24<18:32, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  11%|█▏        | 3861/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3862/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  11%|█▏        | 3862/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3863/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3863/33666 [02:24<18:32, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3864/33666 [02:24<18:32, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3864/33666 [02:24<18:32, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3865/33666 [02:24<18:32, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  11%|█▏        | 3865/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3866/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3866/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3867/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3867/33666 [02:24<18:32, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3868/33666 [02:24<18:32, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3868/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3869/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3869/33666 [02:24<18:32, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3870/33666 [02:24<18:32, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  11%|█▏        | 3870/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3871/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  11%|█▏        | 3871/33666 [02:24<18:32, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3872/33666 [02:24<18:32, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3872/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3873/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3873/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3874/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3874/33666 [02:24<18:32, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3875/33666 [02:24<18:32, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3875/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3876/33666 [02:24<18:32, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3876/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3877/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3877/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3878/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3878/33666 [02:24<18:32, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3879/33666 [02:24<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3879/33666 [02:24<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3880/33666 [02:24<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3880/33666 [02:24<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3881/33666 [02:24<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3881/33666 [02:24<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3882/33666 [02:24<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3882/33666 [02:24<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3883/33666 [02:24<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3883/33666 [02:24<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3884/33666 [02:24<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3884/33666 [02:24<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3885/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3885/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3886/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3886/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3887/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3887/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3888/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3888/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3889/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3889/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3890/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3890/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3891/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3891/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3892/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3892/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3893/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3893/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3894/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3894/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3895/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3895/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3896/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3896/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3897/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3897/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3898/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3898/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3899/33666 [02:25<18:31, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3899/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3900/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3900/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3901/33666 [02:25<18:31, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3901/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3902/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3902/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3903/33666 [02:25<18:31, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3903/33666 [02:25<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3904/33666 [02:25<18:31, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3904/33666 [02:25<18:31, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3905/33666 [02:25<18:30, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3905/33666 [02:25<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3906/33666 [02:25<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3906/33666 [02:25<18:30, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3907/33666 [02:25<18:30, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3907/33666 [02:25<18:30, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3908/33666 [02:25<18:30, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3908/33666 [02:25<18:30, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3909/33666 [02:25<18:30, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3909/33666 [02:25<18:30, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3910/33666 [02:25<18:30, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3910/33666 [02:25<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3911/33666 [02:25<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3911/33666 [02:25<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3912/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3912/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3913/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3913/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3914/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3914/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3915/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3915/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3916/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3916/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3917/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3917/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3918/33666 [02:26<18:30, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3918/33666 [02:26<18:30, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3919/33666 [02:26<18:30, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3919/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3920/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3920/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3921/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3921/33666 [02:26<18:30, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3922/33666 [02:26<18:30, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3922/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3923/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3923/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3924/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3924/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3925/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3925/33666 [02:26<18:30, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3926/33666 [02:26<18:30, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3926/33666 [02:26<18:30, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3927/33666 [02:26<18:30, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3927/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3928/33666 [02:26<18:30, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3928/33666 [02:26<18:30, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3929/33666 [02:26<18:29, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3929/33666 [02:26<18:29, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3930/33666 [02:26<18:29, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3930/33666 [02:26<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3931/33666 [02:26<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3931/33666 [02:26<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3932/33666 [02:26<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3932/33666 [02:26<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3933/33666 [02:26<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3933/33666 [02:26<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3934/33666 [02:26<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3934/33666 [02:26<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3935/33666 [02:26<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3935/33666 [02:26<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3936/33666 [02:26<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3936/33666 [02:26<18:29, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3937/33666 [02:26<18:29, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3937/33666 [02:26<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3938/33666 [02:26<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3938/33666 [02:26<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3939/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3939/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3940/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3940/33666 [02:27<18:29, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3941/33666 [02:27<18:29, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3941/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3942/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3942/33666 [02:27<18:29, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3943/33666 [02:27<18:29, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3943/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3944/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3944/33666 [02:27<18:29, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3945/33666 [02:27<18:29, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3945/33666 [02:27<18:29, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3946/33666 [02:27<18:29, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3946/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3947/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3947/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3948/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3948/33666 [02:27<18:29, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:  12%|█▏        | 3949/33666 [02:27<18:29, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:  12%|█▏        | 3949/33666 [02:27<18:29, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 3950/33666 [02:27<18:29, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 3950/33666 [02:27<18:29, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 3951/33666 [02:27<18:29, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 3951/33666 [02:27<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3952/33666 [02:27<18:29, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3952/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3953/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3953/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3954/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3954/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3955/33666 [02:27<18:29, 26.79it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 3955/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3956/33666 [02:27<18:29, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3956/33666 [02:27<18:29, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3957/33666 [02:27<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3957/33666 [02:27<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3958/33666 [02:27<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3958/33666 [02:27<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3959/33666 [02:27<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3959/33666 [02:27<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3960/33666 [02:27<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3960/33666 [02:27<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3961/33666 [02:27<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3961/33666 [02:27<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3962/33666 [02:27<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3962/33666 [02:27<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3963/33666 [02:27<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3963/33666 [02:27<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3964/33666 [02:27<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3964/33666 [02:27<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3965/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3965/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3966/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3966/33666 [02:28<18:28, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 3967/33666 [02:28<18:28, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 3967/33666 [02:28<18:28, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:  12%|█▏        | 3968/33666 [02:28<18:28, 26.79it/s, loss=4.62, v_num=0]
Epoch 0:  12%|█▏        | 3968/33666 [02:28<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3969/33666 [02:28<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3969/33666 [02:28<18:28, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3970/33666 [02:28<18:28, 26.79it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 3970/33666 [02:28<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3971/33666 [02:28<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3971/33666 [02:28<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3972/33666 [02:28<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3972/33666 [02:28<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3973/33666 [02:28<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3973/33666 [02:28<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3974/33666 [02:28<18:28, 26.79it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 3974/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3975/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3975/33666 [02:28<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3976/33666 [02:28<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3976/33666 [02:28<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3977/33666 [02:28<18:28, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3977/33666 [02:28<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3978/33666 [02:28<18:28, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3978/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3979/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3979/33666 [02:28<18:28, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3980/33666 [02:28<18:27, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3980/33666 [02:28<18:27, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3981/33666 [02:28<18:27, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 3981/33666 [02:28<18:27, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3982/33666 [02:28<18:27, 26.79it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 3982/33666 [02:28<18:27, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3983/33666 [02:28<18:27, 26.79it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 3983/33666 [02:28<18:27, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3984/33666 [02:28<18:27, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3984/33666 [02:28<18:27, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3985/33666 [02:28<18:27, 26.79it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3985/33666 [02:28<18:27, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3986/33666 [02:28<18:27, 26.79it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 3986/33666 [02:28<18:27, 26.79it/s, loss=4.49, v_num=0]
Epoch 0:  12%|█▏        | 3987/33666 [02:28<18:27, 26.79it/s, loss=4.49, v_num=0]
Epoch 0:  12%|█▏        | 3987/33666 [02:28<18:27, 26.79it/s, loss=4.48, v_num=0]
Epoch 0:  12%|█▏        | 3988/33666 [02:28<18:27, 26.79it/s, loss=4.48, v_num=0]
Epoch 0:  12%|█▏        | 3988/33666 [02:28<18:27, 26.79it/s, loss=4.47, v_num=0]
Epoch 0:  12%|█▏        | 3989/33666 [02:28<18:27, 26.80it/s, loss=4.47, v_num=0]
Epoch 0:  12%|█▏        | 3989/33666 [02:28<18:27, 26.80it/s, loss=4.49, v_num=0]
Epoch 0:  12%|█▏        | 3990/33666 [02:28<18:27, 26.80it/s, loss=4.49, v_num=0]
Epoch 0:  12%|█▏        | 3990/33666 [02:28<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 3991/33666 [02:28<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 3991/33666 [02:28<18:27, 26.79it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3992/33666 [02:28<18:27, 26.80it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3992/33666 [02:28<18:27, 26.80it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3993/33666 [02:29<18:27, 26.79it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3993/33666 [02:29<18:27, 26.79it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3994/33666 [02:29<18:27, 26.79it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3994/33666 [02:29<18:27, 26.79it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3995/33666 [02:29<18:27, 26.80it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3995/33666 [02:29<18:27, 26.79it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3996/33666 [02:29<18:27, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 3996/33666 [02:29<18:27, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3997/33666 [02:29<18:27, 26.80it/s, loss=4.53, v_num=0]
Epoch 0:  12%|█▏        | 3997/33666 [02:29<18:27, 26.80it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3998/33666 [02:29<18:27, 26.80it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 3998/33666 [02:29<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 3999/33666 [02:29<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 3999/33666 [02:29<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 4000/33666 [02:29<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 4000/33666 [02:29<18:27, 26.80it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 4001/33666 [02:29<18:27, 26.80it/s, loss=4.51, v_num=0]
Epoch 0:  12%|█▏        | 4001/33666 [02:29<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 4002/33666 [02:29<18:27, 26.80it/s, loss=4.5, v_num=0]
Epoch 0:  12%|█▏        | 4002/33666 [02:29<18:27, 26.80it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 4003/33666 [02:29<18:27, 26.80it/s, loss=4.52, v_num=0]
Epoch 0:  12%|█▏        | 4003/33666 [02:29<18:27, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 4004/33666 [02:29<18:26, 26.80it/s, loss=4.55, v_num=0]
Epoch 0:  12%|█▏        | 4004/33666 [02:29<18:26, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 4005/33666 [02:29<18:26, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 4005/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4006/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4006/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4007/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4007/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4008/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4008/33666 [02:29<18:26, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4009/33666 [02:29<18:26, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4009/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4010/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4010/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4011/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4011/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4012/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4012/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4013/33666 [02:29<18:26, 26.80it/s, loss=4.59, v_num=0]
Epoch 0:  12%|█▏        | 4013/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4014/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4014/33666 [02:29<18:26, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4015/33666 [02:29<18:26, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4015/33666 [02:29<18:26, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 4016/33666 [02:29<18:26, 26.80it/s, loss=4.54, v_num=0]
Epoch 0:  12%|█▏        | 4016/33666 [02:29<18:26, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 4017/33666 [02:29<18:26, 26.80it/s, loss=4.56, v_num=0]
Epoch 0:  12%|█▏        | 4017/33666 [02:29<18:26, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4018/33666 [02:29<18:26, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4018/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4019/33666 [02:29<18:26, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4019/33666 [02:29<18:26, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4020/33666 [02:30<18:26, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4020/33666 [02:30<18:26, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:  12%|█▏        | 4021/33666 [02:30<18:26, 26.80it/s, loss=4.62, v_num=0]
Epoch 0:  12%|█▏        | 4021/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4022/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4022/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4023/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4023/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4024/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4024/33666 [02:30<18:26, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:  12%|█▏        | 4025/33666 [02:30<18:26, 26.80it/s, loss=4.66, v_num=0]
Epoch 0:  12%|█▏        | 4025/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4026/33666 [02:30<18:26, 26.80it/s, loss=4.65, v_num=0]
Epoch 0:  12%|█▏        | 4026/33666 [02:30<18:26, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  12%|█▏        | 4027/33666 [02:30<18:26, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  12%|█▏        | 4027/33666 [02:30<18:26, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  12%|█▏        | 4028/33666 [02:30<18:25, 26.80it/s, loss=4.63, v_num=0]
Epoch 0:  12%|█▏        | 4028/33666 [02:30<18:25, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4029/33666 [02:30<18:25, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4029/33666 [02:30<18:25, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4030/33666 [02:30<18:25, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4030/33666 [02:30<18:25, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4031/33666 [02:30<18:25, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4031/33666 [02:30<18:25, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4032/33666 [02:30<18:25, 26.80it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4032/33666 [02:30<18:25, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4033/33666 [02:30<18:25, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4033/33666 [02:30<18:25, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4034/33666 [02:30<18:25, 26.80it/s, loss=4.57, v_num=0]
Epoch 0:  12%|█▏        | 4034/33666 [02:30<18:25, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4035/33666 [02:30<18:25, 26.80it/s, loss=4.58, v_num=0]
Epoch 0:  12%|█▏        | 4035/33666 [02:30<18:25, 26.80it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4036/33666 [02:30<18:25, 26.79it/s, loss=4.61, v_num=0]
Epoch 0:  12%|█▏        | 4036/33666 [02:30<18:25, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4037/33666 [02:30<18:25, 26.79it/s, loss=4.6, v_num=0]
Epoch 0:  12%|█▏        | 4037/33666 [02:30<18:25, 26.79it/s, l