Cannot train/test with video data

[CodyQ3]

How can I train/test with video data?

When I run python launch.py -c expts/09_ek55_avt.txt -t -g, I obtain the warning "No video_clips present":

[2022-02-08 17:15:53,778][func.train][INFO] - Computing clips...
[2022-02-08 17:15:53,779][func.train][WARNING] - No video_clips present
[2022-02-08 17:15:53,779][func.train][INFO] - Created dataset with 23430 elts

I have downloaded and cropped the videos and setup the same folder structure as in the README file.

Executing hasattr(_dataset, 'video_clips') results in False. How to add video_clips to _dataset to properly execute compute_clips in datasets/data.py?

[Anirudh257]

Even I get this error but the final accuracy seems fine.

[rohitgirdhar]

Yes you can ignore that warning.

[CodyQ3]

When I run python launch.py -c expts/09_ek55_avt.txt -t -g, I obtain the output below.

Why are acc1/action and acc5/action always 0. Can you help me?
I guess that the final results should be acc1/action: 12.5 and acc5/action: 30.1, as reported in Table 4 of the paper.

...

[2022-02-10 17:03:06,587][func.train][WARNING] - Could not init from /home/cody/code/armarx_ws/AVT/DATA/pretrained/TIMM/jx_vit_base_p16_224-80ecf9dd.pth: []
[2022-02-10 17:03:06,588][func.train][WARNING] - Unused keys in /home/cody/code/armarx_ws/AVT/DATA/pretrained/TIMM/jx_vit_base_p16_224-80ecf9dd.pth: ['head.bias', 'head.weight']
[2022-02-10 17:03:07,837][func.train][INFO] - Using LR 0.000100 WD 0.000100 for parameters dict_keys(['__all__.backbone.model.cls_token', '__all__.backbone.model.pos_embed', '__all__.backbone.model.patch_embed.proj.weight', '__all__.backbone.model.blocks.0.norm1.weight', '__all__.backbone.model.blocks.0.attn.qkv.weight', '__all__.backbone.model.blocks.0.attn.proj.weight', '__all__.backbone.model.blocks.0.norm2.weight', '__all__.backbone.model.blocks.0.mlp.fc1.weight', '__all__.backbone.model.blocks.0.mlp.fc2.weight', '__all__.backbone.model.blocks.1.norm1.weight', '__all__.backbone.model.blocks.1.attn.qkv.weight', '__all__.backbone.model.blocks.1.attn.proj.weight', '__all__.backbone.model.blocks.1.norm2.weight', '__all__.backbone.model.blocks.1.mlp.fc1.weight', '__all__.backbone.model.blocks.1.mlp.fc2.weight', '__all__.backbone.model.blocks.2.norm1.weight', '__all__.backbone.model.blocks.2.attn.qkv.weight', '__all__.backbone.model.blocks.2.attn.proj.weight', '__all__.backbone.model.blocks.2.norm2.weight', '__all__.backbone.model.blocks.2.mlp.fc1.weight', '__all__.backbone.model.blocks.2.mlp.fc2.weight', '__all__.backbone.model.blocks.3.norm1.weight', '__all__.backbone.model.blocks.3.attn.qkv.weight', '__all__.backbone.model.blocks.3.attn.proj.weight', '__all__.backbone.model.blocks.3.norm2.weight', '__all__.backbone.model.blocks.3.mlp.fc1.weight', '__all__.backbone.model.blocks.3.mlp.fc2.weight', '__all__.backbone.model.blocks.4.norm1.weight', '__all__.backbone.model.blocks.4.attn.qkv.weight', '__all__.backbone.model.blocks.4.attn.proj.weight', '__all__.backbone.model.blocks.4.norm2.weight', '__all__.backbone.model.blocks.4.mlp.fc1.weight', '__all__.backbone.model.blocks.4.mlp.fc2.weight', '__all__.backbone.model.blocks.5.norm1.weight', '__all__.backbone.model.blocks.5.attn.qkv.weight', '__all__.backbone.model.blocks.5.attn.proj.weight', '__all__.backbone.model.blocks.5.norm2.weight', '__all__.backbone.model.blocks.5.mlp.fc1.weight', '__all__.backbone.model.blocks.5.mlp.fc2.weight', '__all__.backbone.model.blocks.6.norm1.weight', '__all__.backbone.model.blocks.6.attn.qkv.weight', '__all__.backbone.model.blocks.6.attn.proj.weight', '__all__.backbone.model.blocks.6.norm2.weight', '__all__.backbone.model.blocks.6.mlp.fc1.weight', '__all__.backbone.model.blocks.6.mlp.fc2.weight', '__all__.backbone.model.blocks.7.norm1.weight', '__all__.backbone.model.blocks.7.attn.qkv.weight', '__all__.backbone.model.blocks.7.attn.proj.weight', '__all__.backbone.model.blocks.7.norm2.weight', '__all__.backbone.model.blocks.7.mlp.fc1.weight', '__all__.backbone.model.blocks.7.mlp.fc2.weight', '__all__.backbone.model.blocks.8.norm1.weight', '__all__.backbone.model.blocks.8.attn.qkv.weight', '__all__.backbone.model.blocks.8.attn.proj.weight', '__all__.backbone.model.blocks.8.norm2.weight', '__all__.backbone.model.blocks.8.mlp.fc1.weight', '__all__.backbone.model.blocks.8.mlp.fc2.weight', '__all__.backbone.model.blocks.9.norm1.weight', '__all__.backbone.model.blocks.9.attn.qkv.weight', '__all__.backbone.model.blocks.9.attn.proj.weight', '__all__.backbone.model.blocks.9.norm2.weight', '__all__.backbone.model.blocks.9.mlp.fc1.weight', '__all__.backbone.model.blocks.9.mlp.fc2.weight', '__all__.backbone.model.blocks.10.norm1.weight', '__all__.backbone.model.blocks.10.attn.qkv.weight', '__all__.backbone.model.blocks.10.attn.proj.weight', '__all__.backbone.model.blocks.10.norm2.weight', '__all__.backbone.model.blocks.10.mlp.fc1.weight', '__all__.backbone.model.blocks.10.mlp.fc2.weight', '__all__.backbone.model.blocks.11.norm1.weight', '__all__.backbone.model.blocks.11.attn.qkv.weight', '__all__.backbone.model.blocks.11.attn.proj.weight', '__all__.backbone.model.blocks.11.norm2.weight', '__all__.backbone.model.blocks.11.mlp.fc1.weight', '__all__.backbone.model.blocks.11.mlp.fc2.weight', '__all__.backbone.model.norm.weight', '__all__.future_predictor.encoder.weight', '__all__.future_predictor.decoder.weight', '__all__.future_predictor.gpt_model.wpe.weight', '__all__.future_predictor.gpt_model.h.0.ln_1.weight', '__all__.future_predictor.gpt_model.h.0.attn.c_attn.weight', '__all__.future_predictor.gpt_model.h.0.attn.c_proj.weight', '__all__.future_predictor.gpt_model.h.0.ln_2.weight', '__all__.future_predictor.gpt_model.h.0.mlp.c_fc.weight', '__all__.future_predictor.gpt_model.h.0.mlp.c_proj.weight', '__all__.future_predictor.gpt_model.h.1.ln_1.weight', '__all__.future_predictor.gpt_model.h.1.attn.c_attn.weight', '__all__.future_predictor.gpt_model.h.1.attn.c_proj.weight', '__all__.future_predictor.gpt_model.h.1.ln_2.weight', '__all__.future_predictor.gpt_model.h.1.mlp.c_fc.weight', '__all__.future_predictor.gpt_model.h.1.mlp.c_proj.weight', '__all__.future_predictor.gpt_model.h.2.ln_1.weight', '__all__.future_predictor.gpt_model.h.2.attn.c_attn.weight', '__all__.future_predictor.gpt_model.h.2.attn.c_proj.weight', '__all__.future_predictor.gpt_model.h.2.ln_2.weight', '__all__.future_predictor.gpt_model.h.2.mlp.c_fc.weight', '__all__.future_predictor.gpt_model.h.2.mlp.c_proj.weight', '__all__.future_predictor.gpt_model.h.3.ln_1.weight', '__all__.future_predictor.gpt_model.h.3.attn.c_attn.weight', '__all__.future_predictor.gpt_model.h.3.attn.c_proj.weight', '__all__.future_predictor.gpt_model.h.3.ln_2.weight', '__all__.future_predictor.gpt_model.h.3.mlp.c_fc.weight', '__all__.future_predictor.gpt_model.h.3.mlp.c_proj.weight', '__all__.future_predictor.gpt_model.h.4.ln_1.weight', '__all__.future_predictor.gpt_model.h.4.attn.c_attn.weight', '__all__.future_predictor.gpt_model.h.4.attn.c_proj.weight', '__all__.future_predictor.gpt_model.h.4.ln_2.weight', '__all__.future_predictor.gpt_model.h.4.mlp.c_fc.weight', '__all__.future_predictor.gpt_model.h.4.mlp.c_proj.weight', '__all__.future_predictor.gpt_model.h.5.ln_1.weight', '__all__.future_predictor.gpt_model.h.5.attn.c_attn.weight', '__all__.future_predictor.gpt_model.h.5.attn.c_proj.weight', '__all__.future_predictor.gpt_model.h.5.ln_2.weight', '__all__.future_predictor.gpt_model.h.5.mlp.c_fc.weight', '__all__.future_predictor.gpt_model.h.5.mlp.c_proj.weight', '__all__.future_predictor.gpt_model.ln_f.weight', '__all__.classifiers.action.weight'])
[2022-02-10 17:03:07,838][func.train][INFO] - Using LR 0.000100 WD 0.000100 for parameters dict_keys(['__all__.backbone.model.patch_embed.proj.bias', '__all__.backbone.model.blocks.0.norm1.bias', '__all__.backbone.model.blocks.0.attn.qkv.bias', '__all__.backbone.model.blocks.0.attn.proj.bias', '__all__.backbone.model.blocks.0.norm2.bias', '__all__.backbone.model.blocks.0.mlp.fc1.bias', '__all__.backbone.model.blocks.0.mlp.fc2.bias', '__all__.backbone.model.blocks.1.norm1.bias', '__all__.backbone.model.blocks.1.attn.qkv.bias', '__all__.backbone.model.blocks.1.attn.proj.bias', '__all__.backbone.model.blocks.1.norm2.bias', '__all__.backbone.model.blocks.1.mlp.fc1.bias', '__all__.backbone.model.blocks.1.mlp.fc2.bias', '__all__.backbone.model.blocks.2.norm1.bias', '__all__.backbone.model.blocks.2.attn.qkv.bias', '__all__.backbone.model.blocks.2.attn.proj.bias', '__all__.backbone.model.blocks.2.norm2.bias', '__all__.backbone.model.blocks.2.mlp.fc1.bias', '__all__.backbone.model.blocks.2.mlp.fc2.bias', '__all__.backbone.model.blocks.3.norm1.bias', '__all__.backbone.model.blocks.3.attn.qkv.bias', '__all__.backbone.model.blocks.3.attn.proj.bias', '__all__.backbone.model.blocks.3.norm2.bias', '__all__.backbone.model.blocks.3.mlp.fc1.bias', '__all__.backbone.model.blocks.3.mlp.fc2.bias', '__all__.backbone.model.blocks.4.norm1.bias', '__all__.backbone.model.blocks.4.attn.qkv.bias', '__all__.backbone.model.blocks.4.attn.proj.bias', '__all__.backbone.model.blocks.4.norm2.bias', '__all__.backbone.model.blocks.4.mlp.fc1.bias', '__all__.backbone.model.blocks.4.mlp.fc2.bias', '__all__.backbone.model.blocks.5.norm1.bias', '__all__.backbone.model.blocks.5.attn.qkv.bias', '__all__.backbone.model.blocks.5.attn.proj.bias', '__all__.backbone.model.blocks.5.norm2.bias', '__all__.backbone.model.blocks.5.mlp.fc1.bias', '__all__.backbone.model.blocks.5.mlp.fc2.bias', '__all__.backbone.model.blocks.6.norm1.bias', '__all__.backbone.model.blocks.6.attn.qkv.bias', '__all__.backbone.model.blocks.6.attn.proj.bias', '__all__.backbone.model.blocks.6.norm2.bias', '__all__.backbone.model.blocks.6.mlp.fc1.bias', '__all__.backbone.model.blocks.6.mlp.fc2.bias', '__all__.backbone.model.blocks.7.norm1.bias', '__all__.backbone.model.blocks.7.attn.qkv.bias', '__all__.backbone.model.blocks.7.attn.proj.bias', '__all__.backbone.model.blocks.7.norm2.bias', '__all__.backbone.model.blocks.7.mlp.fc1.bias', '__all__.backbone.model.blocks.7.mlp.fc2.bias', '__all__.backbone.model.blocks.8.norm1.bias', '__all__.backbone.model.blocks.8.attn.qkv.bias', '__all__.backbone.model.blocks.8.attn.proj.bias', '__all__.backbone.model.blocks.8.norm2.bias', '__all__.backbone.model.blocks.8.mlp.fc1.bias', '__all__.backbone.model.blocks.8.mlp.fc2.bias', '__all__.backbone.model.blocks.9.norm1.bias', '__all__.backbone.model.blocks.9.attn.qkv.bias', '__all__.backbone.model.blocks.9.attn.proj.bias', '__all__.backbone.model.blocks.9.norm2.bias', '__all__.backbone.model.blocks.9.mlp.fc1.bias', '__all__.backbone.model.blocks.9.mlp.fc2.bias', '__all__.backbone.model.blocks.10.norm1.bias', '__all__.backbone.model.blocks.10.attn.qkv.bias', '__all__.backbone.model.blocks.10.attn.proj.bias', '__all__.backbone.model.blocks.10.norm2.bias', '__all__.backbone.model.blocks.10.mlp.fc1.bias', '__all__.backbone.model.blocks.10.mlp.fc2.bias', '__all__.backbone.model.blocks.11.norm1.bias', '__all__.backbone.model.blocks.11.attn.qkv.bias', '__all__.backbone.model.blocks.11.attn.proj.bias', '__all__.backbone.model.blocks.11.norm2.bias', '__all__.backbone.model.blocks.11.mlp.fc1.bias', '__all__.backbone.model.blocks.11.mlp.fc2.bias', '__all__.backbone.model.norm.bias', '__all__.future_predictor.gpt_model.h.0.ln_1.bias', '__all__.future_predictor.gpt_model.h.0.attn.c_attn.bias', '__all__.future_predictor.gpt_model.h.0.attn.c_proj.bias', '__all__.future_predictor.gpt_model.h.0.ln_2.bias', '__all__.future_predictor.gpt_model.h.0.mlp.c_fc.bias', '__all__.future_predictor.gpt_model.h.0.mlp.c_proj.bias', '__all__.future_predictor.gpt_model.h.1.ln_1.bias', '__all__.future_predictor.gpt_model.h.1.attn.c_attn.bias', '__all__.future_predictor.gpt_model.h.1.attn.c_proj.bias', '__all__.future_predictor.gpt_model.h.1.ln_2.bias', '__all__.future_predictor.gpt_model.h.1.mlp.c_fc.bias', '__all__.future_predictor.gpt_model.h.1.mlp.c_proj.bias', '__all__.future_predictor.gpt_model.h.2.ln_1.bias', '__all__.future_predictor.gpt_model.h.2.attn.c_attn.bias', '__all__.future_predictor.gpt_model.h.2.attn.c_proj.bias', '__all__.future_predictor.gpt_model.h.2.ln_2.bias', '__all__.future_predictor.gpt_model.h.2.mlp.c_fc.bias', '__all__.future_predictor.gpt_model.h.2.mlp.c_proj.bias', '__all__.future_predictor.gpt_model.h.3.ln_1.bias', '__all__.future_predictor.gpt_model.h.3.attn.c_attn.bias', '__all__.future_predictor.gpt_model.h.3.attn.c_proj.bias', '__all__.future_predictor.gpt_model.h.3.ln_2.bias', '__all__.future_predictor.gpt_model.h.3.mlp.c_fc.bias', '__all__.future_predictor.gpt_model.h.3.mlp.c_proj.bias', '__all__.future_predictor.gpt_model.h.4.ln_1.bias', '__all__.future_predictor.gpt_model.h.4.attn.c_attn.bias', '__all__.future_predictor.gpt_model.h.4.attn.c_proj.bias', '__all__.future_predictor.gpt_model.h.4.ln_2.bias', '__all__.future_predictor.gpt_model.h.4.mlp.c_fc.bias', '__all__.future_predictor.gpt_model.h.4.mlp.c_proj.bias', '__all__.future_predictor.gpt_model.h.5.ln_1.bias', '__all__.future_predictor.gpt_model.h.5.attn.c_attn.bias', '__all__.future_predictor.gpt_model.h.5.attn.c_proj.bias', '__all__.future_predictor.gpt_model.h.5.ln_2.bias', '__all__.future_predictor.gpt_model.h.5.mlp.c_fc.bias', '__all__.future_predictor.gpt_model.h.5.mlp.c_proj.bias', '__all__.future_predictor.gpt_model.ln_f.bias', '__all__.classifiers.action.bias'])
[2022-02-10 17:03:07,842][func.train][INFO] - Wrapping model into DDP
[2022-02-10 17:03:08,119][func.train][INFO] - Starting test_only
[2022-02-10 17:03:08,120][func.train][INFO] - Running evaluation for dataset_eval
[2022-02-10 17:03:08,120][func.train][INFO] - Clearing /home/cody/code/armarx_ws/AVT/OUTPUTS/expts/09_ek55_avt.txt/0/results//*
[2022-02-10 17:03:08,579][py.warnings][WARNING] - /home/cody/anaconda3/envs/avt2/lib/python3.7/site-packages/torch/nn/functional.py:3635: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details.
  "See the documentation of nn.Upsample for details.".format(mode)

...

[2022-02-10 18:00:12,013][func.train][INFO] - [] Test:  [1194/1657]  eta: 0:22:06  iter_time: 2.6477  data_time: 1.2725  loss: 14.9744 (14.3621)  acc1/action: 0.0000 (0.0837)  acc5/action: 0.0000 (0.3068)  cls_action: 7.8229 (7.8434)  past_cls_action: 6.0362 (5.4542)  feat: 1.0515 (1.0644)  time: 2.6477  data: 1.2725  max mem: 3444
[2022-02-10 18:00:17,133][func.train][INFO] - [] Test:  [1196/1657]  eta: 0:22:00  iter_time: 2.6395  data_time: 1.2703  loss: 14.6731 (14.3632)  acc1/action: 0.0000 (0.0835)  acc5/action: 0.0000 (0.3063)  cls_action: 7.8154 (7.8434)  past_cls_action: 5.7466 (5.4554)  feat: 1.0515 (1.0644)  time: 2.6395  data: 1.2703  max mem: 3444
[2022-02-10 18:00:22,424][func.train][INFO] - [] Test:  [1198/1657]  eta: 0:21:54  iter_time: 2.6315  data_time: 1.2586  loss: 14.6731 (14.3632)  acc1/action: 0.0000 (0.0834)  acc5/action: 0.0000 (0.3058)  cls_action: 7.7960 (7.8434)  past_cls_action: 5.7466 (5.4554)  feat: 1.0535 (1.0644)  time: 2.6315  data: 1.2586  max mem: 3444
[2022-02-10 18:00:27,805][func.train][INFO] - [] Test:  [1200/1657]  eta: 0:21:48  iter_time: 2.6332  data_time: 1.2626  loss: 14.9199 (14.3647)  acc1/action: 0.0000 (0.0833)  acc5/action: 0.0000 (0.3053)  cls_action: 7.7960 (7.8433)  past_cls_action: 5.9879 (5.4570)  feat: 1.0535 (1.0644)  time: 2.6332  data: 1.2626  max mem: 3444
[2022-02-10 18:00:33,209][func.train][INFO] - [] Test:  [1202/1657]  eta: 0:21:42  iter_time: 2.6338  data_time: 1.2646  loss: 14.6731 (14.3640)  acc1/action: 0.0000 (0.0831)  acc5/action: 0.0000 (0.3048)  cls_action: 7.7904 (7.8433)  past_cls_action: 5.7477 (5.4564)  feat: 1.0517 (1.0643)  time: 2.6338  data: 1.2646  max mem: 3444
[2022-02-10 18:00:38,545][func.train][INFO] - [] Test:  [1204/1657]  eta: 0:21:36  iter_time: 2.6407  data_time: 1.2679  loss: 14.9199 (14.3670)  acc1/action: 0.0000 (0.0830)  acc5/action: 0.0000 (0.3043)  cls_action: 7.7904 (7.8433)  past_cls_action: 5.9879 (5.4594)  feat: 1.0517 (1.0643)  time: 2.6407  data: 1.2679  max mem: 3444
[2022-02-10 18:00:43,647][func.train][INFO] - [] Test:  [1206/1657]  eta: 0:21:30  iter_time: 2.6305  data_time: 1.2574  loss: 14.6915 (14.3651)  acc1/action: 0.0000 (0.0829)  acc5/action: 0.0000 (0.3038)  cls_action: 7.8406 (7.8434)  past_cls_action: 5.7596 (5.4575)  feat: 1.0544 (1.0643)  time: 2.6305  data: 1.2574  max mem: 3444
[2022-02-10 18:00:48,816][func.train][INFO] - [] Test:  [1208/1657]  eta: 0:21:24  iter_time: 2.6245  data_time: 1.2625  loss: 14.6915 (14.3658)  acc1/action: 0.0000 (0.0827)  acc5/action: 0.0000 (0.3033)  cls_action: 7.8222 (7.8433)  past_cls_action: 5.7596 (5.4582)  feat: 1.0517 (1.0642)  time: 2.6245  data: 1.2625  max mem: 3444
[2022-02-10 18:00:54,231][func.train][INFO] - [] Test:  [1210/1657]  eta: 0:21:19  iter_time: 2.6478  data_time: 1.2784  loss: 14.6915 (14.3679)  acc1/action: 0.0000 (0.0826)  acc5/action: 0.0000 (0.3028)  cls_action: 7.8237 (7.8433)  past_cls_action: 5.7596 (5.4604)  feat: 1.0505 (1.0642)  time: 2.6478  data: 1.2784  max mem: 3444
[2022-02-10 18:00:59,537][func.train][INFO] - [] Test:  [1212/1657]  eta: 0:21:13  iter_time: 2.6464  data_time: 1.2765  loss: 15.1471 (14.3709)  acc1/action: 0.0000 (0.0824)  acc5/action: 0.0000 (0.3023)  cls_action: 7.8237 (7.8432)  past_cls_action: 6.2057 (5.4635)  feat: 1.0470 (1.0642)  time: 2.6464  data: 1.2765  max mem: 3444
[2022-02-10 18:01:04,591][func.train][INFO] - [] Test:  [1214/1657]  eta: 0:21:07  iter_time: 2.6282  data_time: 1.2640  loss: 15.1471 (14.3740)  acc1/action: 0.0000 (0.0823)  acc5/action: 0.0000 (0.3018)  cls_action: 7.8406 (7.8433)  past_cls_action: 6.2057 (5.4665)  feat: 1.0452 (1.0641)  time: 2.6282  data: 1.2640  max mem: 3444
[2022-02-10 18:01:09,849][func.train][INFO] - [] Test:  [1216/1657]  eta: 0:21:01  iter_time: 2.6351  data_time: 1.2663  loss: 15.7418 (14.3767)  acc1/action: 0.0000 (0.0822)  acc5/action: 0.0000 (0.3013)  cls_action: 7.8406 (7.8433)  past_cls_action: 6.8394 (5.4693)  feat: 1.0452 (1.0641)  time: 2.6351  data: 1.2663  max mem: 3444
[2022-02-10 18:01:14,805][func.train][INFO] - [] Test:  [1218/1657]  eta: 0:20:55  iter_time: 2.6183  data_time: 1.2658  loss: 15.7619 (14.3782)  acc1/action: 0.0000 (0.0820)  acc5/action: 0.0000 (0.3281)  cls_action: 7.8237 (7.8432)  past_cls_action: 6.8608 (5.4709)  feat: 1.0452 (1.0641)  time: 2.6183  data: 1.2658  max mem: 3444
[2022-02-10 18:01:20,166][func.train][INFO] - [] Test:  [1220/1657]  eta: 0:20:49  iter_time: 2.6173  data_time: 1.2684  loss: 15.7619 (14.3811)  acc1/action: 0.0000 (0.0819)  acc5/action: 0.0000 (0.3276)  cls_action: 7.8406 (7.8432)  past_cls_action: 6.8608 (5.4738)  feat: 1.0497 (1.0641)  time: 2.6173  data: 1.2684  max mem: 3444
[2022-02-10 18:01:25,419][func.train][INFO] - [] Test:  [1222/1657]  eta: 0:20:43  iter_time: 2.6097  data_time: 1.2666  loss: 15.8813 (14.3848)  acc1/action: 0.0000 (0.0818)  acc5/action: 0.0000 (0.3271)  cls_action: 7.8406 (7.8432)  past_cls_action: 6.9623 (5.4775)  feat: 1.0511 (1.0641)  time: 2.6097  data: 1.2666  max mem: 3444
[2022-02-10 18:01:29,584][func.train][INFO] - [] Test:  [1224/1657]  eta: 0:20:37  iter_time: 2.5512  data_time: 1.2122  loss: 15.7619 (14.3814)  acc1/action: 0.0000 (0.0816)  acc5/action: 0.0000 (0.3265)  cls_action: 7.8321 (7.8432)  past_cls_action: 6.8822 (5.4741)  feat: 1.0497 (1.0640)  time: 2.5512  data: 1.2122  max mem: 3444
[2022-02-10 18:01:34,739][func.train][INFO] - [] Test:  [1226/1657]  eta: 0:20:31  iter_time: 2.5538  data_time: 1.2207  loss: 16.1108 (14.3851)  acc1/action: 0.0000 (0.0815)  acc5/action: 0.0000 (0.3260)  cls_action: 7.8222 (7.8431)  past_cls_action: 7.2948 (5.4779)  feat: 1.0511 (1.0640)  time: 2.5538  data: 1.2207  max mem: 3444
[2022-02-10 18:01:40,151][func.train][INFO] - [] Test:  [1228/1657]  eta: 0:20:25  iter_time: 2.5660  data_time: 1.2298  loss: 16.1108 (14.3880)  acc1/action: 0.0000 (0.0814)  acc5/action: 0.0000 (0.3255)  cls_action: 7.8321 (7.8432)  past_cls_action: 7.2948 (5.4809)  feat: 1.0511 (1.0640)  time: 2.5660  data: 1.2298  max mem: 3444
[2022-02-10 18:01:45,380][func.train][INFO] - [] Test:  [1230/1657]  eta: 0:20:19  iter_time: 2.5567  data_time: 1.2245  loss: 15.8853 (14.3881)  acc1/action: 0.0000 (0.0812)  acc5/action: 0.0000 (0.3249)  cls_action: 7.8310 (7.8431)  past_cls_action: 7.0249 (5.4811)  feat: 1.0497 (1.0639)  time: 2.5567  data: 1.2245  max mem: 3444
[2022-02-10 18:01:50,399][func.train][INFO] - [] Test:  [1232/1657]  eta: 0:20:13  iter_time: 2.5423  data_time: 1.2187  loss: 15.8276 (14.3890)  acc1/action: 0.0000 (0.0811)  acc5/action: 0.0000 (0.3244)  cls_action: 7.8321 (7.8431)  past_cls_action: 6.9623 (5.4821)  feat: 1.0497 (1.0639)  time: 2.5423  data: 1.2187  max mem: 3444
[2022-02-10 18:01:55,785][func.train][INFO] - [] Test:  [1234/1657]  eta: 0:20:07  iter_time: 2.5589  data_time: 1.2319  loss: 15.8276 (14.3927)  acc1/action: 0.0000 (0.0810)  acc5/action: 0.0000 (0.3239)  cls_action: 7.8310 (7.8432)  past_cls_action: 6.9818 (5.4857)  feat: 1.0497 (1.0638)  time: 2.5589  data: 1.2319  max mem: 3444
[2022-02-10 18:02:00,907][func.train][INFO] - [] Test:  [1236/1657]  eta: 0:20:02  iter_time: 2.5522  data_time: 1.2358  loss: 15.8853 (14.3954)  acc1/action: 0.0000 (0.0808)  acc5/action: 0.0000 (0.3234)  cls_action: 7.8213 (7.8431)  past_cls_action: 7.0249 (5.4884)  feat: 1.0500 (1.0639)  time: 2.5522  data: 1.2358  max mem: 3444
[2022-02-10 18:02:06,463][func.train][INFO] - [] Test:  [1238/1657]  eta: 0:19:56  iter_time: 2.5821  data_time: 1.2520  loss: 16.1920 (14.3992)  acc1/action: 0.0000 (0.0807)  acc5/action: 0.0000 (0.3228)  cls_action: 7.8310 (7.8431)  past_cls_action: 7.3136 (5.4923)  feat: 1.0473 (1.0638)  time: 2.5821  data: 1.2520  max mem: 3444
[2022-02-10 18:02:11,903][func.train][INFO] - [] Test:  [1240/1657]  eta: 0:19:50  iter_time: 2.5861  data_time: 1.2559  loss: 16.1920 (14.4024)  acc1/action: 0.0000 (0.0806)  acc5/action: 0.0000 (0.3223)  cls_action: 7.8213 (7.8431)  past_cls_action: 7.3136 (5.4956)  feat: 1.0434 (1.0638)  time: 2.5861  data: 1.2559  max mem: 3444
[2022-02-10 18:02:17,166][func.train][INFO] - [] Test:  [1242/1657]  eta: 0:19:44  iter_time: 2.5866  data_time: 1.2614  loss: 16.0963 (14.4048)  acc1/action: 0.0000 (0.0805)  acc5/action: 0.0000 (0.3218)  cls_action: 7.8213 (7.8431)  past_cls_action: 7.2753 (5.4980)  feat: 1.0434 (1.0637)  time: 2.5866  data: 1.2614  max mem: 3444
[2022-02-10 18:02:22,598][func.train][INFO] - [] Test:  [1244/1657]  eta: 0:19:38  iter_time: 2.6500  data_time: 1.3220  loss: 16.0963 (14.4049)  acc1/action: 0.0000 (0.0803)  acc5/action: 0.0000 (0.3213)  cls_action: 7.8310 (7.8431)  past_cls_action: 7.2753 (5.4982)  feat: 1.0434 (1.0637)  time: 2.6500  data: 1.3220  max mem: 3444
[2022-02-10 18:02:27,882][func.train][INFO] - [] Test:  [1246/1657]  eta: 0:19:32  iter_time: 2.6564  data_time: 1.3290  loss: 16.0963 (14.4083)  acc1/action: 0.0000 (0.0802)  acc5/action: 0.0000 (0.3208)  cls_action: 7.8332 (7.8430)  past_cls_action: 7.2753 (5.5016)  feat: 1.0434 (1.0637)  time: 2.6564  data: 1.3290  max mem: 3444
[2022-02-10 18:02:33,207][func.train][INFO] - [] Test:  [1248/1657]  eta: 0:19:27  iter_time: 2.6521  data_time: 1.3267  loss: 15.9462 (14.4095)  acc1/action: 0.0000 (0.0801)  acc5/action: 0.0000 (0.3203)  cls_action: 7.8310 (7.8430)  past_cls_action: 7.0706 (5.5028)  feat: 1.0436 (1.0637)  time: 2.6521  data: 1.3267  max mem: 3444
[2022-02-10 18:02:38,480][func.train][INFO] - [] Test:  [1250/1657]  eta: 0:19:21  iter_time: 2.6543  data_time: 1.3308  loss: 16.0963 (14.4125)  acc1/action: 0.0000 (0.0799)  acc5/action: 0.0000 (0.3197)  cls_action: 7.8348 (7.8429)  past_cls_action: 7.2753 (5.5059)  feat: 1.0450 (1.0636)  time: 2.6543  data: 1.3308  max mem: 3444
[2022-02-10 18:02:43,951][func.train][INFO] - [] Test:  [1252/1657]  eta: 0:19:15  iter_time: 2.6769  data_time: 1.3465  loss: 16.1920 (14.4156)  acc1/action: 0.0000 (0.0798)  acc5/action: 0.0000 (0.3192)  cls_action: 7.8354 (7.8430)  past_cls_action: 7.3136 (5.5089)  feat: 1.0450 (1.0636)  time: 2.6769  data: 1.3465  max mem: 3444
[2022-02-10 18:02:49,455][func.train][INFO] - [] Test:  [1254/1657]  eta: 0:19:09  iter_time: 2.6828  data_time: 1.3563  loss: 16.0963 (14.4171)  acc1/action: 0.0000 (0.0797)  acc5/action: 0.0000 (0.3187)  cls_action: 7.8407 (7.8431)  past_cls_action: 7.2753 (5.5104)  feat: 1.0450 (1.0636)  time: 2.6828  data: 1.3563  max mem: 3444
[2022-02-10 18:02:54,653][func.train][INFO] - [] Test:  [1256/1657]  eta: 0:19:03  iter_time: 2.6866  data_time: 1.3595  loss: 16.0465 (14.4171)  acc1/action: 0.0000 (0.0796)  acc5/action: 0.0000 (0.3182)  cls_action: 7.8423 (7.8431)  past_cls_action: 7.2636 (5.5104)  feat: 1.0449 (1.0635)  time: 2.6866  data: 1.3595  max mem: 3444
[2022-02-10 18:03:00,186][func.train][INFO] - [] Test:  [1258/1657]  eta: 0:18:58  iter_time: 2.6855  data_time: 1.3654  loss: 15.6305 (14.4147)  acc1/action: 0.0000 (0.0794)  acc5/action: 0.0000 (0.3177)  cls_action: 7.8423 (7.8431)  past_cls_action: 6.7508 (5.5081)  feat: 1.0450 (1.0635)  time: 2.6855  data: 1.3654  max mem: 3444
[2022-02-10 18:03:05,696][func.train][INFO] - [] Test:  [1260/1657]  eta: 0:18:52  iter_time: 2.6890  data_time: 1.3681  loss: 15.1354 (14.4139)  acc1/action: 0.0000 (0.0793)  acc5/action: 0.0000 (0.3172)  cls_action: 7.8437 (7.8431)  past_cls_action: 6.2804 (5.5072)  feat: 1.0459 (1.0635)  time: 2.6890  data: 1.3681  max mem: 3444
[2022-02-10 18:03:11,123][func.train][INFO] - [] Test:  [1262/1657]  eta: 0:18:46  iter_time: 2.6972  data_time: 1.3744  loss: 14.8546 (14.4139)  acc1/action: 0.0000 (0.0792)  acc5/action: 0.0000 (0.3167)  cls_action: 7.8472 (7.8431)  past_cls_action: 5.9726 (5.5074)  feat: 1.0459 (1.0635)  time: 2.6972  data: 1.3744  max mem: 3444
[2022-02-10 18:03:16,419][func.train][INFO] - [] Test:  [1264/1657]  eta: 0:18:40  iter_time: 2.6904  data_time: 1.3670  loss: 14.7760 (14.4114)  acc1/action: 0.0000 (0.0791)  acc5/action: 0.0000 (0.3162)  cls_action: 7.8480 (7.8431)  past_cls_action: 5.8355 (5.5049)  feat: 1.0459 (1.0634)  time: 2.6904  data: 1.3670  max mem: 3444
[2022-02-10 18:03:21,895][func.train][INFO] - [] Test:  [1266/1657]  eta: 0:18:34  iter_time: 2.7000  data_time: 1.3770  loss: 14.6778 (14.4118)  acc1/action: 0.0000 (0.0789)  acc5/action: 0.0000 (0.3157)  cls_action: 7.8495 (7.8431)  past_cls_action: 5.7795 (5.5053)  feat: 1.0447 (1.0634)  time: 2.7000  data: 1.3770  max mem: 3444
[2022-02-10 18:03:27,193][func.train][INFO] - [] Test:  [1268/1657]  eta: 0:18:29  iter_time: 2.6986  data_time: 1.3817  loss: 14.6778 (14.4144)  acc1/action: 0.0000 (0.0788)  acc5/action: 0.0000 (0.3152)  cls_action: 7.8478 (7.8430)  past_cls_action: 5.7795 (5.5081)  feat: 1.0410 (1.0633)  time: 2.6986  data: 1.3817  max mem: 3444
[2022-02-10 18:03:32,462][func.train][INFO] - [] Test:  [1270/1657]  eta: 0:18:23  iter_time: 2.6985  data_time: 1.3797  loss: 14.6778 (14.4168)  acc1/action: 0.0000 (0.0787)  acc5/action: 0.0000 (0.3147)  cls_action: 7.8472 (7.8429)  past_cls_action: 5.7795 (5.5107)  feat: 1.0398 (1.0632)  time: 2.6985  data: 1.3797  max mem: 3444
[2022-02-10 18:03:37,660][func.train][INFO] - [] Test:  [1272/1657]  eta: 0:18:17  iter_time: 2.6848  data_time: 1.3739  loss: 14.6425 (14.4178)  acc1/action: 0.0000 (0.0786)  acc5/action: 0.0000 (0.3142)  cls_action: 7.8437 (7.8429)  past_cls_action: 5.7768 (5.5117)  feat: 1.0398 (1.0632)  time: 2.6848  data: 1.3739  max mem: 3444
[2022-02-10 18:03:43,006][func.train][INFO] - [] Test:  [1274/1657]  eta: 0:18:11  iter_time: 2.6769  data_time: 1.3614  loss: 14.6103 (14.4187)  acc1/action: 0.0000 (0.0784)  acc5/action: 0.0000 (0.3137)  cls_action: 7.8223 (7.8428)  past_cls_action: 5.7549 (5.5127)  feat: 1.0398 (1.0632)  time: 2.6769  data: 1.3614  max mem: 3444
[2022-02-10 18:03:48,268][func.train][INFO] - [] Test:  [1276/1657]  eta: 0:18:05  iter_time: 2.6800  data_time: 1.3619  loss: 14.5835 (14.4180)  acc1/action: 0.0000 (0.0783)  acc5/action: 0.0000 (0.3132)  cls_action: 7.8404 (7.8429)  past_cls_action: 5.7071 (5.5120)  feat: 1.0410 (1.0632)  time: 2.6800  data: 1.3619  max mem: 3444
[2022-02-10 18:03:53,742][func.train][INFO] - [] Test:  [1278/1657]  eta: 0:18:00  iter_time: 2.6771  data_time: 1.3565  loss: 14.6425 (14.4200)  acc1/action: 0.0000 (0.0782)  acc5/action: 0.0000 (0.3127)  cls_action: 7.8404 (7.8428)  past_cls_action: 5.7656 (5.5139)  feat: 1.0418 (1.0632)  time: 2.6771  data: 1.3565  max mem: 3444
[2022-02-10 18:03:58,871][func.train][INFO] - [] Test:  [1280/1657]  eta: 0:17:54  iter_time: 2.6580  data_time: 1.3455  loss: 14.8546 (14.4230)  acc1/action: 0.0000 (0.0781)  acc5/action: 0.0000 (0.3123)  cls_action: 7.8223 (7.8428)  past_cls_action: 5.9726 (5.5171)  feat: 1.0432 (1.0632)  time: 2.6580  data: 1.3455  max mem: 3444
[2022-02-10 18:04:04,388][func.train][INFO] - [] Test:  [1282/1657]  eta: 0:17:48  iter_time: 2.6625  data_time: 1.3474  loss: 15.0593 (14.4248)  acc1/action: 0.0000 (0.0779)  acc5/action: 0.0000 (0.3118)  cls_action: 7.8223 (7.8427)  past_cls_action: 6.2283 (5.5190)  feat: 1.0444 (1.0632)  time: 2.6625  data: 1.3474  max mem: 3444
[2022-02-10 18:04:09,794][func.train][INFO] - [] Test:  [1284/1657]  eta: 0:17:42  iter_time: 2.6680  data_time: 1.3591  loss: 15.4875 (14.4269)  acc1/action: 0.0000 (0.0778)  acc5/action: 0.0000 (0.3113)  cls_action: 7.8157 (7.8426)  past_cls_action: 6.5365 (5.5211)  feat: 1.0508 (1.0632)  time: 2.6680  data: 1.3591  max mem: 3444
[2022-02-10 18:04:15,278][func.train][INFO] - [] Test:  [1286/1657]  eta: 0:17:36  iter_time: 2.6684  data_time: 1.3510  loss: 15.7307 (14.4297)  acc1/action: 0.0000 (0.0777)  acc5/action: 0.0000 (0.3108)  cls_action: 7.7847 (7.8425)  past_cls_action: 7.0127 (5.5241)  feat: 1.0535 (1.0631)  time: 2.6684  data: 1.3510  max mem: 3444
[2022-02-10 18:04:20,721][func.train][INFO] - [] Test:  [1288/1657]  eta: 0:17:31  iter_time: 2.6757  data_time: 1.3493  loss: 15.7307 (14.4327)  acc1/action: 0.0000 (0.0776)  acc5/action: 0.0000 (0.3103)  cls_action: 7.8157 (7.8424)  past_cls_action: 7.0127 (5.5271)  feat: 1.0540 (1.0631)  time: 2.6757  data: 1.3493  max mem: 3444
[2022-02-10 18:04:26,137][func.train][INFO] - [] Test:  [1290/1657]  eta: 0:17:25  iter_time: 2.6830  data_time: 1.3577  loss: 15.9370 (14.4358)  acc1/action: 0.0000 (0.0775)  acc5/action: 0.0000 (0.3098)  cls_action: 7.7895 (7.8423)  past_cls_action: 7.1558 (5.5303)  feat: 1.0547 (1.0631)  time: 2.6830  data: 1.3577  max mem: 3444
[2022-02-10 18:04:31,390][func.train][INFO] - [] Test:  [1292/1657]  eta: 0:17:19  iter_time: 2.6858  data_time: 1.3596  loss: 16.2315 (14.4391)  acc1/action: 0.0000 (0.0773)  acc5/action: 0.0000 (0.3094)  cls_action: 7.8157 (7.8423)  past_cls_action: 7.4269 (5.5337)  feat: 1.0547 (1.0631)  time: 2.6858  data: 1.3596  max mem: 3444
[2022-02-10 18:04:36,814][func.train][INFO] - [] Test:  [1294/1657]  eta: 0:17:13  iter_time: 2.6897  data_time: 1.3687  loss: 16.2315 (14.4404)  acc1/action: 0.0000 (0.0772)  acc5/action: 0.0000 (0.3089)  cls_action: 7.8172 (7.8424)  past_cls_action: 7.4269 (5.5349)  feat: 1.0553 (1.0631)  time: 2.6897  data: 1.3687  max mem: 3444
[2022-02-10 18:04:42,119][func.train][INFO] - [] Test:  [1296/1657]  eta: 0:17:07  iter_time: 2.6918  data_time: 1.3667  loss: 16.2315 (14.4422)  acc1/action: 0.0000 (0.0771)  acc5/action: 0.0000 (0.3084)  cls_action: 7.8157 (7.8424)  past_cls_action: 7.4269 (5.5366)  feat: 1.0562 (1.0631)  time: 2.6918  data: 1.3667  max mem: 3444
[2022-02-10 18:04:47,347][func.train][INFO] - [] Test:  [1298/1657]  eta: 0:17:02  iter_time: 2.6796  data_time: 1.3624  loss: 16.1450 (14.4438)  acc1/action: 0.0000 (0.0770)  acc5/action: 0.0000 (0.3079)  cls_action: 7.8157 (7.8425)  past_cls_action: 7.2480 (5.5382)  feat: 1.0553 (1.0631)  time: 2.6796  data: 1.3624  max mem: 3444
[2022-02-10 18:04:52,580][func.train][INFO] - [] Test:  [1300/1657]  eta: 0:16:56  iter_time: 2.6848  data_time: 1.3638  loss: 15.9370 (14.4462)  acc1/action: 0.0000 (0.0769)  acc5/action: 0.0000 (0.3075)  cls_action: 7.7895 (7.8424)  past_cls_action: 7.1558 (5.5407)  feat: 1.0547 (1.0630)  time: 2.6848  data: 1.3638  max mem: 3444
[2022-02-10 18:04:57,913][func.train][INFO] - [] Test:  [1302/1657]  eta: 0:16:50  iter_time: 2.6756  data_time: 1.3543  loss: 15.8684 (14.4478)  acc1/action: 0.0000 (0.0767)  acc5/action: 0.0000 (0.3070)  cls_action: 7.7895 (7.8423)  past_cls_action: 7.0451 (5.5425)  feat: 1.0519 (1.0630)  time: 2.6756  data: 1.3543  max mem: 3444
[2022-02-10 18:05:03,258][func.train][INFO] - [] Test:  [1304/1657]  eta: 0:16:44  iter_time: 2.6725  data_time: 1.3437  loss: 15.8529 (14.4489)  acc1/action: 0.0000 (0.0766)  acc5/action: 0.0000 (0.3065)  cls_action: 7.8045 (7.8423)  past_cls_action: 6.9833 (5.5436)  feat: 1.0495 (1.0630)  time: 2.6725  data: 1.3437  max mem: 3444
[2022-02-10 18:05:08,378][func.train][INFO] - [] Test:  [1306/1657]  eta: 0:16:38  iter_time: 2.6543  data_time: 1.3238  loss: 15.7477 (14.4475)  acc1/action: 0.0000 (0.0765)  acc5/action: 0.0000 (0.3060)  cls_action: 7.8379 (7.8423)  past_cls_action: 6.8226 (5.5422)  feat: 1.0495 (1.0630)  time: 2.6543  data: 1.3238  max mem: 3444
[2022-02-10 18:05:13,885][func.train][INFO] - [] Test:  [1308/1657]  eta: 0:16:33  iter_time: 2.6575  data_time: 1.3275  loss: 15.4798 (14.4447)  acc1/action: 0.0000 (0.0764)  acc5/action: 0.0000 (0.3056)  cls_action: 7.8356 (7.8422)  past_cls_action: 6.5189 (5.5396)  feat: 1.0495 (1.0629)  time: 2.6575  data: 1.3275  max mem: 3444
[2022-02-10 18:05:19,087][func.train][INFO] - [] Test:  [1310/1657]  eta: 0:16:27  iter_time: 2.6468  data_time: 1.3234  loss: 15.3150 (14.4466)  acc1/action: 0.0000 (0.0763)  acc5/action: 0.0000 (0.3051)  cls_action: 7.8527 (7.8422)  past_cls_action: 6.4836 (5.5415)  feat: 1.0495 (1.0629)  time: 2.6468  data: 1.3234  max mem: 3444
[2022-02-10 18:05:24,304][func.train][INFO] - [] Test:  [1312/1657]  eta: 0:16:21  iter_time: 2.6450  data_time: 1.3204  loss: 15.2539 (14.4436)  acc1/action: 0.0000 (0.0762)  acc5/action: 0.0000 (0.3046)  cls_action: 7.8527 (7.8423)  past_cls_action: 6.3295 (5.5384)  feat: 1.0452 (1.0629)  time: 2.6450  data: 1.3204  max mem: 3444
[2022-02-10 18:05:29,606][func.train][INFO] - [] Test:  [1314/1657]  eta: 0:16:15  iter_time: 2.6389  data_time: 1.3222  loss: 15.1731 (14.4413)  acc1/action: 0.0000 (0.0760)  acc5/action: 0.0000 (0.3042)  cls_action: 7.8356 (7.8423)  past_cls_action: 6.2788 (5.5362)  feat: 1.0425 (1.0628)  time: 2.6389  data: 1.3222  max mem: 3444
[2022-02-10 18:05:35,030][func.train][INFO] - [] Test:  [1316/1657]  eta: 0:16:09  iter_time: 2.6449  data_time: 1.3234  loss: 15.0726 (14.4395)  acc1/action: 0.0000 (0.0759)  acc5/action: 0.0000 (0.3037)  cls_action: 7.8356 (7.8423)  past_cls_action: 6.1986 (5.5345)  feat: 1.0416 (1.0628)  time: 2.6449  data: 1.3234  max mem: 3444
[2022-02-10 18:05:40,078][func.train][INFO] - [] Test:  [1318/1657]  eta: 0:16:04  iter_time: 2.6358  data_time: 1.3144  loss: 14.4008 (14.4357)  acc1/action: 0.0000 (0.0758)  acc5/action: 0.0000 (0.3033)  cls_action: 7.8335 (7.8423)  past_cls_action: 5.6257 (5.5307)  feat: 1.0407 (1.0627)  time: 2.6358  data: 1.3144  max mem: 3444
[2022-02-10 18:05:44,833][func.train][INFO] - [] Test:  [1320/1657]  eta: 0:15:58  iter_time: 2.6119  data_time: 1.2844  loss: 12.8816 (14.4319)  acc1/action: 0.0000 (0.0757)  acc5/action: 0.0000 (0.3028)  cls_action: 7.8356 (7.8423)  past_cls_action: 3.9513 (5.5269)  feat: 1.0416 (1.0627)  time: 2.6119  data: 1.2844  max mem: 3444
[2022-02-10 18:05:50,866][func.train][INFO] - [] Test:  [1322/1657]  eta: 0:15:52  iter_time: 2.6469  data_time: 1.3197  loss: 12.8816 (14.4337)  acc1/action: 0.0000 (0.0756)  acc5/action: 0.0000 (0.3023)  cls_action: 7.8527 (7.8423)  past_cls_action: 3.9513 (5.5287)  feat: 1.0416 (1.0627)  time: 2.6469  data: 1.3197  max mem: 3444
[2022-02-10 18:05:56,811][func.train][INFO] - [] Test:  [1324/1657]  eta: 0:15:46  iter_time: 2.6770  data_time: 1.3515  loss: 12.8816 (14.4355)  acc1/action: 0.0000 (0.0755)  acc5/action: 0.0000 (0.3019)  cls_action: 7.8550 (7.8423)  past_cls_action: 3.9513 (5.5306)  feat: 1.0407 (1.0626)  time: 2.6770  data: 1.3515  max mem: 3444
[2022-02-10 18:06:02,750][func.train][INFO] - [] Test:  [1326/1657]  eta: 0:15:41  iter_time: 2.7179  data_time: 1.4011  loss: 13.1330 (14.4344)  acc1/action: 0.0000 (0.0754)  acc5/action: 0.0000 (0.3014)  cls_action: 7.8550 (7.8424)  past_cls_action: 4.1801 (5.5294)  feat: 1.0407 (1.0626)  time: 2.7179  data: 1.4011  max mem: 3444
[2022-02-10 18:06:08,750][func.train][INFO] - [] Test:  [1328/1657]  eta: 0:15:35  iter_time: 2.7425  data_time: 1.4281  loss: 13.5717 (14.4346)  acc1/action: 0.0000 (0.0752)  acc5/action: 0.0000 (0.3010)  cls_action: 7.8550 (7.8424)  past_cls_action: 4.6978 (5.5295)  feat: 1.0407 (1.0626)  time: 2.7425  data: 1.4281  max mem: 3444
[2022-02-10 18:06:14,822][func.train][INFO] - [] Test:  [1330/1657]  eta: 0:15:30  iter_time: 2.7861  data_time: 1.4748  loss: 13.1330 (14.4330)  acc1/action: 0.0000 (0.0751)  acc5/action: 0.0000 (0.3005)  cls_action: 7.8550 (7.8424)  past_cls_action: 4.1942 (5.5279)  feat: 1.0407 (1.0626)  time: 2.7861  data: 1.4748  max mem: 3444
[2022-02-10 18:06:20,859][func.train][INFO] - [] Test:  [1332/1657]  eta: 0:15:24  iter_time: 2.8271  data_time: 1.5114  loss: 13.5805 (14.4327)  acc1/action: 0.0000 (0.0750)  acc5/action: 0.0000 (0.3001)  cls_action: 7.8453 (7.8424)  past_cls_action: 4.7015 (5.5277)  feat: 1.0451 (1.0626)  time: 2.8271  data: 1.5114  max mem: 3444
[2022-02-10 18:06:26,486][func.train][INFO] - [] Test:  [1334/1657]  eta: 0:15:18  iter_time: 2.8433  data_time: 1.5261  loss: 13.5805 (14.4289)  acc1/action: 0.0000 (0.0749)  acc5/action: 0.0000 (0.2996)  cls_action: 7.8453 (7.8423)  past_cls_action: 4.7015 (5.5239)  feat: 1.0498 (1.0627)  time: 2.8433  data: 1.5261  max mem: 3444
[2022-02-10 18:06:32,373][func.train][INFO] - [] Test:  [1336/1657]  eta: 0:15:13  iter_time: 2.8665  data_time: 1.5615  loss: 13.5805 (14.4264)  acc1/action: 0.0000 (0.0748)  acc5/action: 0.0000 (0.2992)  cls_action: 7.8453 (7.8422)  past_cls_action: 4.7015 (5.5215)  feat: 1.0563 (1.0627)  time: 2.8665  data: 1.5615  max mem: 3444
[2022-02-10 18:06:38,637][func.train][INFO] - [] Test:  [1338/1657]  eta: 0:15:07  iter_time: 2.9273  data_time: 1.6171  loss: 13.6065 (14.4240)  acc1/action: 0.0000 (0.0747)  acc5/action: 0.0000 (0.2987)  cls_action: 7.8487 (7.8423)  past_cls_action: 4.7015 (5.5191)  feat: 1.0563 (1.0626)  time: 2.9273  data: 1.6171  max mem: 3444
[2022-02-10 18:06:44,590][func.train][INFO] - [] Test:  [1340/1657]  eta: 0:15:01  iter_time: 2.9872  data_time: 1.6874  loss: 13.6506 (14.4224)  acc1/action: 0.0000 (0.0746)  acc5/action: 0.0000 (0.2983)  cls_action: 7.8487 (7.8423)  past_cls_action: 4.7325 (5.5175)  feat: 1.0563 (1.0626)  time: 2.9872  data: 1.6874  max mem: 3444
[2022-02-10 18:06:50,284][func.train][INFO] - [] Test:  [1342/1657]  eta: 0:14:56  iter_time: 2.9702  data_time: 1.6749  loss: 13.6506 (14.4245)  acc1/action: 0.0000 (0.0745)  acc5/action: 0.0000 (0.2978)  cls_action: 7.8453 (7.8422)  past_cls_action: 4.7325 (5.5197)  feat: 1.0603 (1.0626)  time: 2.9702  data: 1.6749  max mem: 3444
[2022-02-10 18:06:56,104][func.train][INFO] - [] Test:  [1344/1657]  eta: 0:14:50  iter_time: 2.9640  data_time: 1.6712  loss: 13.6506 (14.4275)  acc1/action: 0.0000 (0.0743)  acc5/action: 0.0000 (0.2974)  cls_action: 7.8487 (7.8423)  past_cls_action: 4.7325 (5.5227)  feat: 1.0624 (1.0626)  time: 2.9640  data: 1.6712  max mem: 3444
[2022-02-10 18:07:02,242][func.train][INFO] - [] Test:  [1346/1657]  eta: 0:14:44  iter_time: 2.9739  data_time: 1.6742  loss: 13.7412 (14.4278)  acc1/action: 0.0000 (0.0742)  acc5/action: 0.0000 (0.2970)  cls_action: 7.8487 (7.8424)  past_cls_action: 4.9261 (5.5229)  feat: 1.0624 (1.0626)  time: 2.9739  data: 1.6742  max mem: 3444
[2022-02-10 18:07:07,955][func.train][INFO] - [] Test:  [1348/1657]  eta: 0:14:39  iter_time: 2.9595  data_time: 1.6680  loss: 13.7412 (14.4284)  acc1/action: 0.0000 (0.0741)  acc5/action: 0.0000 (0.2965)  cls_action: 7.8487 (7.8422)  past_cls_action: 4.9261 (5.5236)  feat: 1.0624 (1.0626)  time: 2.9595  data: 1.6680  max mem: 3444
[2022-02-10 18:07:13,794][func.train][INFO] - [] Test:  [1350/1657]  eta: 0:14:33  iter_time: 2.9479  data_time: 1.6466  loss: 13.8122 (14.4286)  acc1/action: 0.0000 (0.0740)  acc5/action: 0.0000 (0.2961)  cls_action: 7.8085 (7.8420)  past_cls_action: 4.9421 (5.5240)  feat: 1.0624 (1.0626)  time: 2.9479  data: 1.6466  max mem: 3444
[2022-02-10 18:07:19,767][func.train][INFO] - [] Test:  [1352/1657]  eta: 0:14:28  iter_time: 2.9446  data_time: 1.6507  loss: 13.8613 (14.4315)  acc1/action: 0.0000 (0.0739)  acc5/action: 0.0000 (0.2956)  cls_action: 7.8590 (7.8420)  past_cls_action: 4.9560 (5.5269)  feat: 1.0644 (1.0626)  time: 2.9446  data: 1.6507  max mem: 3444
[2022-02-10 18:07:25,830][func.train][INFO] - [] Test:  [1354/1657]  eta: 0:14:22  iter_time: 2.9665  data_time: 1.6656  loss: 13.9072 (14.4316)  acc1/action: 0.0000 (0.0738)  acc5/action: 0.0000 (0.2952)  cls_action: 7.8626 (7.8421)  past_cls_action: 4.9826 (5.5269)  feat: 1.0604 (1.0626)  time: 2.9665  data: 1.6656  max mem: 3444
[2022-02-10 18:07:31,940][func.train][INFO] - [] Test:  [1356/1657]  eta: 0:14:16  iter_time: 2.9776  data_time: 1.6690  loss: 15.0725 (14.4313)  acc1/action: 0.0000 (0.0737)  acc5/action: 0.0000 (0.2948)  cls_action: 7.8626 (7.8421)  past_cls_action: 6.2656 (5.5266)  feat: 1.0644 (1.0626)  time: 2.9776  data: 1.6690  max mem: 3444
[2022-02-10 18:07:38,060][func.train][INFO] - [] Test:  [1358/1657]  eta: 0:14:11  iter_time: 2.9704  data_time: 1.6688  loss: 15.0725 (14.4312)  acc1/action: 0.0000 (0.0736)  acc5/action: 0.0000 (0.2943)  cls_action: 7.8351 (7.8420)  past_cls_action: 6.2656 (5.5265)  feat: 1.0645 (1.0627)  time: 2.9704  data: 1.6688  max mem: 3444
[2022-02-10 18:07:44,101][func.train][INFO] - [] Test:  [1360/1657]  eta: 0:14:05  iter_time: 2.9748  data_time: 1.6658  loss: 15.0725 (14.4314)  acc1/action: 0.0000 (0.0735)  acc5/action: 0.0000 (0.2939)  cls_action: 7.7902 (7.8419)  past_cls_action: 6.2656 (5.5267)  feat: 1.0649 (1.0627)  time: 2.9748  data: 1.6658  max mem: 3444
[2022-02-10 18:07:50,179][func.train][INFO] - [] Test:  [1362/1657]  eta: 0:13:59  iter_time: 2.9940  data_time: 1.6881  loss: 14.6420 (14.4294)  acc1/action: 0.0000 (0.0734)  acc5/action: 0.0000 (0.2935)  cls_action: 7.8351 (7.8420)  past_cls_action: 5.7370 (5.5248)  feat: 1.0672 (1.0627)  time: 2.9940  data: 1.6881  max mem: 3444
[2022-02-10 18:07:56,300][func.train][INFO] - [] Test:  [1364/1657]  eta: 0:13:54  iter_time: 3.0090  data_time: 1.7036  loss: 14.5702 (14.4310)  acc1/action: 0.0000 (0.0733)  acc5/action: 0.0000 (0.2930)  cls_action: 7.8351 (7.8420)  past_cls_action: 5.7188 (5.5263)  feat: 1.0672 (1.0627)  time: 3.0090  data: 1.7036  max mem: 3444
[2022-02-10 18:08:02,261][func.train][INFO] - [] Test:  [1366/1657]  eta: 0:13:48  iter_time: 3.0002  data_time: 1.6965  loss: 14.5702 (14.4296)  acc1/action: 0.0000 (0.0732)  acc5/action: 0.0000 (0.2926)  cls_action: 7.8228 (7.8420)  past_cls_action: 5.7188 (5.5249)  feat: 1.0702 (1.0627)  time: 3.0002  data: 1.6965  max mem: 3444
[2022-02-10 18:08:08,361][func.train][INFO] - [] Test:  [1368/1657]  eta: 0:13:43  iter_time: 3.0196  data_time: 1.7152  loss: 14.6420 (14.4328)  acc1/action: 0.0000 (0.0730)  acc5/action: 0.0000 (0.2922)  cls_action: 7.8253 (7.8420)  past_cls_action: 5.7370 (5.5282)  feat: 1.0702 (1.0627)  time: 3.0196  data: 1.7152  max mem: 3444
[2022-02-10 18:08:14,230][func.train][INFO] - [] Test:  [1370/1657]  eta: 0:13:37  iter_time: 3.0211  data_time: 1.7291  loss: 14.6420 (14.4342)  acc1/action: 0.0000 (0.0729)  acc5/action: 0.0000 (0.2918)  cls_action: 7.8354 (7.8420)  past_cls_action: 5.7370 (5.5294)  feat: 1.0675 (1.0627)  time: 3.0211  data: 1.7291  max mem: 3444
[2022-02-10 18:08:20,325][func.train][INFO] - [] Test:  [1372/1657]  eta: 0:13:31  iter_time: 3.0272  data_time: 1.7339  loss: 14.5495 (14.4308)  acc1/action: 0.0000 (0.0728)  acc5/action: 0.0000 (0.2913)  cls_action: 7.8351 (7.8420)  past_cls_action: 5.6944 (5.5261)  feat: 1.0675 (1.0627)  time: 3.0272  data: 1.7339  max mem: 3444
[2022-02-10 18:08:26,399][func.train][INFO] - [] Test:  [1374/1657]  eta: 0:13:26  iter_time: 3.0277  data_time: 1.7344  loss: 14.5495 (14.4309)  acc1/action: 0.0000 (0.0727)  acc5/action: 0.0000 (0.2909)  cls_action: 7.8302 (7.8420)  past_cls_action: 5.6944 (5.5262)  feat: 1.0667 (1.0627)  time: 3.0277  data: 1.7344  max mem: 3444
[2022-02-10 18:08:32,402][func.train][INFO] - [] Test:  [1376/1657]  eta: 0:13:20  iter_time: 3.0224  data_time: 1.7296  loss: 14.5495 (14.4309)  acc1/action: 0.0000 (0.0726)  acc5/action: 0.0000 (0.2905)  cls_action: 7.8316 (7.8420)  past_cls_action: 5.6944 (5.5262)  feat: 1.0666 (1.0627)  time: 3.0224  data: 1.7296  max mem: 3444
[2022-02-10 18:08:38,448][func.train][INFO] - [] Test:  [1378/1657]  eta: 0:13:14  iter_time: 3.0187  data_time: 1.7267  loss: 14.4600 (14.4294)  acc1/action: 0.0000 (0.0725)  acc5/action: 0.0000 (0.2901)  cls_action: 7.8316 (7.8420)  past_cls_action: 5.4872 (5.5246)  feat: 1.0675 (1.0627)  time: 3.0187  data: 1.7267  max mem: 3444
[2022-02-10 18:08:43,996][func.train][INFO] - [] Test:  [1380/1657]  eta: 0:13:09  iter_time: 2.9941  data_time: 1.7094  loss: 14.1928 (14.4294)  acc1/action: 0.0000 (0.0724)  acc5/action: 0.0000 (0.2896)  cls_action: 7.8364 (7.8421)  past_cls_action: 5.2309 (5.5246)  feat: 1.0689 (1.0628)  time: 2.9941  data: 1.7094  max mem: 3444
[2022-02-10 18:08:50,272][func.train][INFO] - [] Test:  [1382/1657]  eta: 0:13:03  iter_time: 3.0039  data_time: 1.7127  loss: 14.4600 (14.4317)  acc1/action: 0.0000 (0.0723)  acc5/action: 0.0000 (0.2892)  cls_action: 7.8364 (7.8420)  past_cls_action: 5.4872 (5.5269)  feat: 1.0675 (1.0628)  time: 3.0039  data: 1.7127  max mem: 3444
[2022-02-10 18:08:56,270][func.train][INFO] - [] Test:  [1384/1657]  eta: 0:12:58  iter_time: 2.9978  data_time: 1.7091  loss: 14.4600 (14.4317)  acc1/action: 0.0000 (0.0722)  acc5/action: 0.0000 (0.2888)  cls_action: 7.8302 (7.8420)  past_cls_action: 5.4872 (5.5270)  feat: 1.0666 (1.0627)  time: 2.9978  data: 1.7091  max mem: 3444
[2022-02-10 18:09:02,336][func.train][INFO] - [] Test:  [1386/1657]  eta: 0:12:52  iter_time: 3.0031  data_time: 1.7137  loss: 14.4600 (14.4328)  acc1/action: 0.0000 (0.0721)  acc5/action: 0.0000 (0.2884)  cls_action: 7.8316 (7.8421)  past_cls_action: 5.4872 (5.5279)  feat: 1.0666 (1.0628)  time: 3.0031  data: 1.7137  max mem: 3444
[2022-02-10 18:09:08,179][func.train][INFO] - [] Test:  [1388/1657]  eta: 0:12:46  iter_time: 2.9902  data_time: 1.6996  loss: 14.1928 (14.4328)  acc1/action: 0.0000 (0.0720)  acc5/action: 0.0000 (0.2880)  cls_action: 7.8364 (7.8420)  past_cls_action: 5.2309 (5.5279)  feat: 1.0666 (1.0628)  time: 2.9902  data: 1.6996  max mem: 3444
[2022-02-10 18:09:14,103][func.train][INFO] - [] Test:  [1390/1657]  eta: 0:12:41  iter_time: 2.9929  data_time: 1.6908  loss: 14.1928 (14.4341)  acc1/action: 0.0000 (0.0719)  acc5/action: 0.0000 (0.2876)  cls_action: 7.8302 (7.8420)  past_cls_action: 5.2309 (5.5293)  feat: 1.0654 (1.0628)  time: 2.9929  data: 1.6908  max mem: 3444
[2022-02-10 18:09:19,832][func.train][INFO] - [] Test:  [1392/1657]  eta: 0:12:35  iter_time: 2.9747  data_time: 1.6723  loss: 14.3055 (14.4313)  acc1/action: 0.0000 (0.0718)  acc5/action: 0.0000 (0.2872)  cls_action: 7.8224 (7.8419)  past_cls_action: 5.4115 (5.5266)  feat: 1.0730 (1.0628)  time: 2.9747  data: 1.6723  max mem: 3444
[2022-02-10 18:09:25,748][func.train][INFO] - [] Test:  [1394/1657]  eta: 0:12:29  iter_time: 2.9668  data_time: 1.6725  loss: 14.3086 (14.4323)  acc1/action: 0.0000 (0.0717)  acc5/action: 0.0000 (0.2867)  cls_action: 7.8267 (7.8420)  past_cls_action: 5.4115 (5.5274)  feat: 1.0745 (1.0629)  time: 2.9668  data: 1.6725  max mem: 3444
[2022-02-10 18:09:32,097][func.train][INFO] - [] Test:  [1396/1657]  eta: 0:12:24  iter_time: 2.9841  data_time: 1.6901  loss: 14.3086 (14.4321)  acc1/action: 0.0000 (0.0716)  acc5/action: 0.0000 (0.2863)  cls_action: 7.8267 (7.8421)  past_cls_action: 5.2826 (5.5270)  feat: 1.0797 (1.0629)  time: 2.9841  data: 1.6901  max mem: 3444
[2022-02-10 18:09:38,185][func.train][INFO] - [] Test:  [1398/1657]  eta: 0:12:18  iter_time: 2.9862  data_time: 1.6840  loss: 14.3098 (14.4309)  acc1/action: 0.0000 (0.0715)  acc5/action: 0.0000 (0.2859)  cls_action: 7.8465 (7.8424)  past_cls_action: 5.4115 (5.5256)  feat: 1.0798 (1.0629)  time: 2.9862  data: 1.6840  max mem: 3444
[2022-02-10 18:09:44,100][func.train][INFO] - [] Test:  [1400/1657]  eta: 0:12:12  iter_time: 3.0045  data_time: 1.7023  loss: 14.3098 (14.4307)  acc1/action: 0.0000 (0.0714)  acc5/action: 0.0000 (0.2855)  cls_action: 7.8465 (7.8423)  past_cls_action: 5.4115 (5.5254)  feat: 1.0745 (1.0630)  time: 3.0045  data: 1.7023  max mem: 3444
[2022-02-10 18:09:49,934][func.train][INFO] - [] Test:  [1402/1657]  eta: 0:12:07  iter_time: 2.9824  data_time: 1.6797  loss: 14.3055 (14.4286)  acc1/action: 0.0000 (0.0713)  acc5/action: 0.0000 (0.2851)  cls_action: 7.8490 (7.8423)  past_cls_action: 5.2726 (5.5233)  feat: 1.0745 (1.0630)  time: 2.9824  data: 1.6797  max mem: 3444
[2022-02-10 18:09:55,902][func.train][INFO] - [] Test:  [1404/1657]  eta: 0:12:01  iter_time: 2.9808  data_time: 1.6795  loss: 14.3086 (14.4303)  acc1/action: 0.0000 (0.0712)  acc5/action: 0.0000 (0.2847)  cls_action: 7.8490 (7.8423)  past_cls_action: 5.2826 (5.5250)  feat: 1.0745 (1.0629)  time: 2.9808  data: 1.6795  max mem: 3444
[2022-02-10 18:10:01,895][func.train][INFO] - [] Test:  [1406/1657]  eta: 0:11:55  iter_time: 2.9772  data_time: 1.6736  loss: 14.3098 (14.4332)  acc1/action: 0.0000 (0.0711)  acc5/action: 0.0000 (0.2843)  cls_action: 7.8398 (7.8424)  past_cls_action: 5.4115 (5.5279)  feat: 1.0700 (1.0630)  time: 2.9772  data: 1.6736  max mem: 3444
[2022-02-10 18:10:07,764][func.train][INFO] - [] Test:  [1408/1657]  eta: 0:11:50  iter_time: 2.9785  data_time: 1.6764  loss: 14.3086 (14.4299)  acc1/action: 0.0000 (0.0710)  acc5/action: 0.0000 (0.2839)  cls_action: 7.8398 (7.8423)  past_cls_action: 5.2826 (5.5247)  feat: 1.0696 (1.0630)  time: 2.9785  data: 1.6764  max mem: 3444
[2022-02-10 18:10:13,804][func.train][INFO] - [] Test:  [1410/1657]  eta: 0:11:44  iter_time: 2.9844  data_time: 1.6807  loss: 14.2639 (14.4266)  acc1/action: 0.0000 (0.0709)  acc5/action: 0.0000 (0.2835)  cls_action: 7.8526 (7.8423)  past_cls_action: 5.2484 (5.5213)  feat: 1.0672 (1.0629)  time: 2.9844  data: 1.6807  max mem: 3444
[2022-02-10 18:10:19,966][func.train][INFO] - [] Test:  [1412/1657]  eta: 0:11:38  iter_time: 3.0059  data_time: 1.6995  loss: 14.3086 (14.4272)  acc1/action: 0.0000 (0.0708)  acc5/action: 0.0000 (0.2831)  cls_action: 7.8603 (7.8423)  past_cls_action: 5.2726 (5.5219)  feat: 1.0672 (1.0630)  time: 3.0059  data: 1.6995  max mem: 3444
[2022-02-10 18:10:26,028][func.train][INFO] - [] Test:  [1414/1657]  eta: 0:11:33  iter_time: 3.0133  data_time: 1.7011  loss: 13.2123 (14.4234)  acc1/action: 0.0000 (0.0707)  acc5/action: 0.0000 (0.2827)  cls_action: 7.8603 (7.8423)  past_cls_action: 4.4155 (5.5180)  feat: 1.0672 (1.0630)  time: 3.0133  data: 1.7011  max mem: 3444
[2022-02-10 18:10:36,351][func.train][INFO] - [] Test:  [1416/1657]  eta: 0:11:28  iter_time: 3.2120  data_time: 1.6840  loss: 13.0298 (14.4175)  acc1/action: 0.0000 (0.0706)  acc5/action: 0.0000 (0.2823)  cls_action: 7.8379 (7.8423)  past_cls_action: 4.1811 (5.5121)  feat: 1.0696 (1.0631)  time: 3.2120  data: 1.6840  max mem: 3444
[2022-02-10 18:10:42,201][func.train][INFO] - [] Test:  [1418/1657]  eta: 0:11:22  iter_time: 3.2000  data_time: 1.6780  loss: 13.0298 (14.4139)  acc1/action: 0.0000 (0.0705)  acc5/action: 0.0000 (0.2819)  cls_action: 7.8360 (7.8422)  past_cls_action: 4.1811 (5.5086)  feat: 1.0696 (1.0631)  time: 3.2000  data: 1.6780  max mem: 3444
[2022-02-10 18:10:48,282][func.train][INFO] - [] Test:  [1420/1657]  eta: 0:11:16  iter_time: 3.2084  data_time: 1.6822  loss: 13.0298 (14.4101)  acc1/action: 0.0000 (0.0704)  acc5/action: 0.0000 (0.2815)  cls_action: 7.8360 (7.8422)  past_cls_action: 4.1811 (5.5047)  feat: 1.0700 (1.0632)  time: 3.2084  data: 1.6822  max mem: 3444
[2022-02-10 18:10:54,319][func.train][INFO] - [] Test:  [1422/1657]  eta: 0:11:11  iter_time: 3.2185  data_time: 1.6978  loss: 13.2123 (14.4098)  acc1/action: 0.0000 (0.0703)  acc5/action: 0.0000 (0.2811)  cls_action: 7.8360 (7.8422)  past_cls_action: 4.4155 (5.5044)  feat: 1.0820 (1.0632)  time: 3.2185  data: 1.6978  max mem: 3444
[2022-02-10 18:11:00,526][func.train][INFO] - [] Test:  [1424/1657]  eta: 0:11:05  iter_time: 3.2305  data_time: 1.7001  loss: 12.2651 (14.4051)  acc1/action: 0.0000 (0.0702)  acc5/action: 0.0000 (0.2807)  cls_action: 7.8244 (7.8421)  past_cls_action: 3.3845 (5.4998)  feat: 1.0820 (1.0633)  time: 3.2305  data: 1.7001  max mem: 3444
[2022-02-10 18:11:06,777][func.train][INFO] - [] Test:  [1426/1657]  eta: 0:11:00  iter_time: 3.2434  data_time: 1.7186  loss: 12.2651 (14.4046)  acc1/action: 0.0000 (0.0701)  acc5/action: 0.0000 (0.2803)  cls_action: 7.8342 (7.8421)  past_cls_action: 3.3845 (5.4992)  feat: 1.0820 (1.0633)  time: 3.2434  data: 1.7186  max mem: 3444
[2022-02-10 18:11:12,863][func.train][INFO] - [] Test:  [1428/1657]  eta: 0:10:54  iter_time: 3.2542  data_time: 1.7184  loss: 13.0843 (14.4062)  acc1/action: 0.0000 (0.0700)  acc5/action: 0.0000 (0.2799)  cls_action: 7.8360 (7.8423)  past_cls_action: 4.2039 (5.5006)  feat: 1.0850 (1.0633)  time: 3.2542  data: 1.7184  max mem: 3444
[2022-02-10 18:11:18,927][func.train][INFO] - [] Test:  [1430/1657]  eta: 0:10:48  iter_time: 3.2554  data_time: 1.7250  loss: 13.4544 (14.4092)  acc1/action: 0.0000 (0.0699)  acc5/action: 0.0000 (0.2795)  cls_action: 7.8342 (7.8423)  past_cls_action: 4.4821 (5.5036)  feat: 1.0871 (1.0634)  time: 3.2554  data: 1.7250  max mem: 3444
[2022-02-10 18:11:24,953][func.train][INFO] - [] Test:  [1432/1657]  eta: 0:10:43  iter_time: 3.2487  data_time: 1.7116  loss: 13.4544 (14.4118)  acc1/action: 0.0000 (0.0698)  acc5/action: 0.0000 (0.2791)  cls_action: 7.8360 (7.8424)  past_cls_action: 4.4821 (5.5059)  feat: 1.0882 (1.0634)  time: 3.2487  data: 1.7116  max mem: 3444
[2022-02-10 18:11:30,994][func.train][INFO] - [] Test:  [1434/1657]  eta: 0:10:37  iter_time: 3.2476  data_time: 1.7104  loss: 13.7493 (14.4115)  acc1/action: 0.0000 (0.0697)  acc5/action: 0.0000 (0.2787)  cls_action: 7.8360 (7.8424)  past_cls_action: 4.8803 (5.5057)  feat: 1.0955 (1.0634)  time: 3.2476  data: 1.7104  max mem: 3444
[2022-02-10 18:11:37,108][func.train][INFO] - [] Test:  [1436/1657]  eta: 0:10:31  iter_time: 3.0372  data_time: 1.7095  loss: 14.1494 (14.4126)  acc1/action: 0.0000 (0.0696)  acc5/action: 0.0000 (0.2784)  cls_action: 7.8433 (7.8425)  past_cls_action: 5.2392 (5.5066)  feat: 1.0915 (1.0635)  time: 3.0372  data: 1.7095  max mem: 3444
[2022-02-10 18:11:43,504][func.train][INFO] - [] Test:  [1438/1657]  eta: 0:10:26  iter_time: 3.0645  data_time: 1.7283  loss: 14.1494 (14.4108)  acc1/action: 0.0000 (0.0695)  acc5/action: 0.0000 (0.2780)  cls_action: 7.8433 (7.8425)  past_cls_action: 5.2392 (5.5048)  feat: 1.0913 (1.0635)  time: 3.0645  data: 1.7283  max mem: 3444
[2022-02-10 18:11:48,627][func.train][INFO] - [] Test:  [1440/1657]  eta: 0:10:20  iter_time: 3.0165  data_time: 1.6712  loss: 14.1494 (14.4072)  acc1/action: 0.0000 (0.0694)  acc5/action: 0.0000 (0.2776)  cls_action: 7.8599 (7.8425)  past_cls_action: 5.2392 (5.5012)  feat: 1.0871 (1.0635)  time: 3.0165  data: 1.6712  max mem: 3444
[2022-02-10 18:11:53,912][func.train][INFO] - [] Test:  [1442/1657]  eta: 0:10:14  iter_time: 2.9789  data_time: 1.6182  loss: 13.8998 (14.4053)  acc1/action: 0.0000 (0.0693)  acc5/action: 0.0000 (0.2772)  cls_action: 7.8610 (7.8426)  past_cls_action: 5.0443 (5.4992)  feat: 1.0800 (1.0635)  time: 2.9789  data: 1.6182  max mem: 3444
[2022-02-10 18:11:59,049][func.train][INFO] - [] Test:  [1444/1657]  eta: 0:10:08  iter_time: 2.9254  data_time: 1.5688  loss: 13.8998 (14.4034)  acc1/action: 0.0000 (0.0692)  acc5/action: 0.0000 (0.2768)  cls_action: 7.8610 (7.8426)  past_cls_action: 5.0443 (5.4974)  feat: 1.0800 (1.0634)  time: 2.9254  data: 1.5688  max mem: 3444
[2022-02-10 18:12:04,286][func.train][INFO] - [] Test:  [1446/1657]  eta: 0:10:02  iter_time: 2.8748  data_time: 1.5070  loss: 13.8998 (14.4015)  acc1/action: 0.0000 (0.0691)  acc5/action: 0.0000 (0.2764)  cls_action: 7.8610 (7.8426)  past_cls_action: 5.0443 (5.4955)  feat: 1.0779 (1.0634)  time: 2.8748  data: 1.5070  max mem: 3444
[2022-02-10 18:12:09,345][func.train][INFO] - [] Test:  [1448/1657]  eta: 0:09:57  iter_time: 2.8234  data_time: 1.4602  loss: 13.7760 (14.3981)  acc1/action: 0.0000 (0.0690)  acc5/action: 0.0000 (0.2761)  cls_action: 7.8555 (7.8425)  past_cls_action: 4.7639 (5.4921)  feat: 1.0747 (1.0634)  time: 2.8234  data: 1.4602  max mem: 3444
[2022-02-10 18:12:14,650][func.train][INFO] - [] Test:  [1450/1657]  eta: 0:09:51  iter_time: 2.7854  data_time: 1.4092  loss: 13.5541 (14.3961)  acc1/action: 0.0000 (0.0689)  acc5/action: 0.0000 (0.2757)  cls_action: 7.8555 (7.8426)  past_cls_action: 4.6876 (5.4901)  feat: 1.0619 (1.0634)  time: 2.7854  data: 1.4092  max mem: 3444
[2022-02-10 18:12:19,697][func.train][INFO] - [] Test:  [1452/1657]  eta: 0:09:45  iter_time: 2.7365  data_time: 1.3656  loss: 12.3680 (14.3920)  acc1/action: 0.0000 (0.0688)  acc5/action: 0.0000 (0.2753)  cls_action: 7.8442 (7.8425)  past_cls_action: 3.4795 (5.4861)  feat: 1.0563 (1.0634)  time: 2.7365  data: 1.3656  max mem: 3444
[2022-02-10 18:12:24,953][func.train][INFO] - [] Test:  [1454/1657]  eta: 0:09:39  iter_time: 2.6972  data_time: 1.3228  loss: 12.3632 (14.3893)  acc1/action: 0.0000 (0.0687)  acc5/action: 0.0000 (0.2749)  cls_action: 7.8549 (7.8425)  past_cls_action: 3.4349 (5.4834)  feat: 1.0511 (1.0633)  time: 2.6972  data: 1.3228  max mem: 3444
[2022-02-10 18:12:30,216][func.train][INFO] - [] Test:  [1456/1657]  eta: 0:09:34  iter_time: 2.6546  data_time: 1.2862  loss: 12.3632 (14.3872)  acc1/action: 0.0000 (0.0686)  acc5/action: 0.0000 (0.2745)  cls_action: 7.8549 (7.8426)  past_cls_action: 3.4349 (5.4813)  feat: 1.0509 (1.0633)  time: 2.6546  data: 1.2862  max mem: 3444
[2022-02-10 18:12:34,828][func.train][INFO] - [] Test:  [1458/1657]  eta: 0:09:28  iter_time: 2.5655  data_time: 1.2072  loss: 12.1994 (14.3813)  acc1/action: 0.0000 (0.0685)  acc5/action: 0.0000 (0.2742)  cls_action: 7.8610 (7.8428)  past_cls_action: 3.3882 (5.4752)  feat: 1.0505 (1.0633)  time: 2.5655  data: 1.2072  max mem: 3444
[2022-02-10 18:12:39,957][func.train][INFO] - [] Test:  [1460/1657]  eta: 0:09:22  iter_time: 2.5658  data_time: 1.2147  loss: 12.3680 (14.3822)  acc1/action: 0.0000 (0.0684)  acc5/action: 0.0000 (0.2738)  cls_action: 7.8549 (7.8428)  past_cls_action: 3.4795 (5.4761)  feat: 1.0493 (1.0633)  time: 2.5658  data: 1.2147  max mem: 3444
[2022-02-10 18:12:45,125][func.train][INFO] - [] Test:  [1462/1657]  eta: 0:09:16  iter_time: 2.5600  data_time: 1.2224  loss: 12.8682 (14.3835)  acc1/action: 0.0000 (0.0684)  acc5/action: 0.0000 (0.2734)  cls_action: 7.8549 (7.8429)  past_cls_action: 3.9351 (5.4774)  feat: 1.0473 (1.0632)  time: 2.5600  data: 1.2224  max mem: 3444
[2022-02-10 18:12:50,475][func.train][INFO] - [] Test:  [1464/1657]  eta: 0:09:10  iter_time: 2.5706  data_time: 1.2296  loss: 12.8682 (14.3834)  acc1/action: 0.0000 (0.0683)  acc5/action: 0.0000 (0.2730)  cls_action: 7.8630 (7.8430)  past_cls_action: 3.9351 (5.4773)  feat: 1.0473 (1.0632)  time: 2.5706  data: 1.2296  max mem: 3444
[2022-02-10 18:12:55,719][func.train][INFO] - [] Test:  [1466/1657]  eta: 0:09:05  iter_time: 2.5709  data_time: 1.2420  loss: 12.8682 (14.3808)  acc1/action: 0.0000 (0.0682)  acc5/action: 0.0000 (0.2727)  cls_action: 7.8685 (7.8430)  past_cls_action: 3.9351 (5.4746)  feat: 1.0457 (1.0631)  time: 2.5709  data: 1.2420  max mem: 3444
[2022-02-10 18:13:01,026][func.train][INFO] - [] Test:  [1468/1657]  eta: 0:08:59  iter_time: 2.5834  data_time: 1.2480  loss: 12.8682 (14.3811)  acc1/action: 0.0000 (0.0681)  acc5/action: 0.0000 (0.2723)  cls_action: 7.8706 (7.8430)  past_cls_action: 3.9351 (5.4749)  feat: 1.0457 (1.0632)  time: 2.5834  data: 1.2480  max mem: 3444
[2022-02-10 18:13:06,138][func.train][INFO] - [] Test:  [1470/1657]  eta: 0:08:53  iter_time: 2.5738  data_time: 1.2540  loss: 12.7508 (14.3778)  acc1/action: 0.0000 (0.0680)  acc5/action: 0.0000 (0.2719)  cls_action: 7.8706 (7.8431)  past_cls_action: 3.7130 (5.4716)  feat: 1.0443 (1.0631)  time: 2.5738  data: 1.2540  max mem: 3444
[2022-02-10 18:13:11,348][func.train][INFO] - [] Test:  [1472/1657]  eta: 0:08:47  iter_time: 2.5819  data_time: 1.2599  loss: 12.8682 (14.3749)  acc1/action: 0.0000 (0.0679)  acc5/action: 0.0000 (0.2716)  cls_action: 7.8706 (7.8430)  past_cls_action: 3.9351 (5.4688)  feat: 1.0402 (1.0631)  time: 2.5819  data: 1.2599  max mem: 3444
[2022-02-10 18:13:16,561][func.train][INFO] - [] Test:  [1474/1657]  eta: 0:08:41  iter_time: 2.5798  data_time: 1.2573  loss: 12.8682 (14.3725)  acc1/action: 0.0000 (0.0678)  acc5/action: 0.0000 (0.2712)  cls_action: 7.8685 (7.8430)  past_cls_action: 3.9351 (5.4665)  feat: 1.0402 (1.0630)  time: 2.5798  data: 1.2573  max mem: 3444
[2022-02-10 18:13:21,590][func.train][INFO] - [] Test:  [1476/1657]  eta: 0:08:36  iter_time: 2.5681  data_time: 1.2477  loss: 12.5546 (14.3679)  acc1/action: 0.0000 (0.0677)  acc5/action: 0.0000 (0.2708)  cls_action: 7.8706 (7.8431)  past_cls_action: 3.7000 (5.4618)  feat: 1.0402 (1.0630)  time: 2.5681  data: 1.2477  max mem: 3444
[2022-02-10 18:13:26,687][func.train][INFO] - [] Test:  [1478/1657]  eta: 0:08:30  iter_time: 2.5923  data_time: 1.2640  loss: 12.5546 (14.3618)  acc1/action: 0.0000 (0.0676)  acc5/action: 0.0000 (0.2705)  cls_action: 7.8706 (7.8432)  past_cls_action: 3.7000 (5.4557)  feat: 1.0385 (1.0630)  time: 2.5923  data: 1.2640  max mem: 3444
[2022-02-10 18:13:31,910][func.train][INFO] - [] Test:  [1480/1657]  eta: 0:08:24  iter_time: 2.5970  data_time: 1.2676  loss: 12.5412 (14.3600)  acc1/action: 0.0000 (0.0675)  acc5/action: 0.0000 (0.2701)  cls_action: 7.8630 (7.8431)  past_cls_action: 3.6609 (5.4539)  feat: 1.0402 (1.0629)  time: 2.5970  data: 1.2676  max mem: 3444
[2022-02-10 18:13:36,935][func.train][INFO] - [] Test:  [1482/1657]  eta: 0:08:18  iter_time: 2.5899  data_time: 1.2648  loss: 12.2927 (14.3579)  acc1/action: 0.0000 (0.0674)  acc5/action: 0.0000 (0.2697)  cls_action: 7.8581 (7.8431)  past_cls_action: 3.4173 (5.4519)  feat: 1.0403 (1.0629)  time: 2.5899  data: 1.2648  max mem: 3444
[2022-02-10 18:13:42,096][func.train][INFO] - [] Test:  [1484/1657]  eta: 0:08:13  iter_time: 2.5804  data_time: 1.2577  loss: 12.1153 (14.3551)  acc1/action: 0.0000 (0.0673)  acc5/action: 0.0000 (0.2694)  cls_action: 7.8564 (7.8431)  past_cls_action: 3.1722 (5.4491)  feat: 1.0403 (1.0629)  time: 2.5804  data: 1.2577  max mem: 3444
[2022-02-10 18:13:51,260][func.train][INFO] - [] Test:  [1486/1657]  eta: 0:08:07  iter_time: 2.7764  data_time: 1.2503  loss: 12.1153 (14.3534)  acc1/action: 0.0000 (0.0672)  acc5/action: 0.0000 (0.2690)  cls_action: 7.8564 (7.8431)  past_cls_action: 3.1722 (5.4474)  feat: 1.0418 (1.0629)  time: 2.7764  data: 1.2503  max mem: 3444
[2022-02-10 18:13:56,421][func.train][INFO] - [] Test:  [1488/1657]  eta: 0:08:02  iter_time: 2.7691  data_time: 1.2551  loss: 12.1153 (14.3540)  acc1/action: 0.0000 (0.0672)  acc5/action: 0.0000 (0.2686)  cls_action: 7.8564 (7.8431)  past_cls_action: 3.1722 (5.4480)  feat: 1.0418 (1.0629)  time: 2.7691  data: 1.2551  max mem: 3444
[2022-02-10 18:14:00,753][func.train][INFO] - [] Test:  [1490/1657]  eta: 0:07:56  iter_time: 2.7301  data_time: 1.2073  loss: 12.1153 (14.3514)  acc1/action: 0.0000 (0.0671)  acc5/action: 0.0000 (0.2683)  cls_action: 7.8564 (7.8431)  past_cls_action: 3.1722 (5.4455)  feat: 1.0406 (1.0628)  time: 2.7301  data: 1.2073  max mem: 3444
[2022-02-10 18:14:05,915][func.train][INFO] - [] Test:  [1492/1657]  eta: 0:07:50  iter_time: 2.7277  data_time: 1.2042  loss: 12.2927 (14.3539)  acc1/action: 0.0000 (0.0670)  acc5/action: 0.0000 (0.2679)  cls_action: 7.8648 (7.8432)  past_cls_action: 3.4173 (5.4480)  feat: 1.0449 (1.0628)  time: 2.7277  data: 1.2042  max mem: 3444
[2022-02-10 18:14:10,930][func.train][INFO] - [] Test:  [1494/1657]  eta: 0:07:44  iter_time: 2.7178  data_time: 1.2025  loss: 13.0643 (14.3562)  acc1/action: 0.0000 (0.0669)  acc5/action: 0.0000 (0.2676)  cls_action: 7.8648 (7.8431)  past_cls_action: 4.1933 (5.4503)  feat: 1.0463 (1.0627)  time: 2.7178  data: 1.2025  max mem: 3444
[2022-02-10 18:14:15,928][func.train][INFO] - [] Test:  [1496/1657]  eta: 0:07:38  iter_time: 2.7163  data_time: 1.1993  loss: 13.5089 (14.3585)  acc1/action: 0.0000 (0.0668)  acc5/action: 0.0000 (0.2672)  cls_action: 7.8648 (7.8432)  past_cls_action: 4.7148 (5.4527)  feat: 1.0463 (1.0627)  time: 2.7163  data: 1.1993  max mem: 3444
[2022-02-10 18:14:20,631][func.train][INFO] - [] Test:  [1498/1657]  eta: 0:07:33  iter_time: 2.6965  data_time: 1.1860  loss: 13.6751 (14.3574)  acc1/action: 0.0000 (0.0667)  acc5/action: 0.0000 (0.2668)  cls_action: 7.8564 (7.8432)  past_cls_action: 4.9179 (5.4515)  feat: 1.0468 (1.0627)  time: 2.6965  data: 1.1860  max mem: 3444
[2022-02-10 18:14:25,667][func.train][INFO] - [] Test:  [1500/1657]  eta: 0:07:27  iter_time: 2.6872  data_time: 1.1764  loss: 14.1917 (14.3598)  acc1/action: 0.0000 (0.0666)  acc5/action: 0.0000 (0.2665)  cls_action: 7.8564 (7.8432)  past_cls_action: 5.2671 (5.4539)  feat: 1.0473 (1.0627)  time: 2.6872  data: 1.1764  max mem: 3444
[2022-02-10 18:14:30,776][func.train][INFO] - [] Test:  [1502/1657]  eta: 0:07:21  iter_time: 2.6914  data_time: 1.1738  loss: 14.1917 (14.3584)  acc1/action: 0.0000 (0.0665)  acc5/action: 0.0000 (0.2661)  cls_action: 7.8648 (7.8432)  past_cls_action: 5.2671 (5.4525)  feat: 1.0479 (1.0627)  time: 2.6914  data: 1.1738  max mem: 3444
[2022-02-10 18:14:36,239][func.train][INFO] - [] Test:  [1504/1657]  eta: 0:07:15  iter_time: 2.7065  data_time: 1.1864  loss: 14.7723 (14.3573)  acc1/action: 0.0000 (0.0664)  acc5/action: 0.0000 (0.2658)  cls_action: 7.8649 (7.8434)  past_cls_action: 5.8010 (5.4512)  feat: 1.0506 (1.0627)  time: 2.7065  data: 1.1864  max mem: 3444
[2022-02-10 18:14:41,365][func.train][INFO] - [] Test:  [1506/1657]  eta: 0:07:10  iter_time: 2.5045  data_time: 1.1899  loss: 14.1917 (14.3555)  acc1/action: 0.0000 (0.0664)  acc5/action: 0.0000 (0.2654)  cls_action: 7.8649 (7.8434)  past_cls_action: 5.2671 (5.4494)  feat: 1.0479 (1.0627)  time: 2.5045  data: 1.1899  max mem: 3444
[2022-02-10 18:14:46,202][func.train][INFO] - [] Test:  [1508/1657]  eta: 0:07:04  iter_time: 2.4883  data_time: 1.1657  loss: 14.1906 (14.3512)  acc1/action: 0.0000 (0.0663)  acc5/action: 0.0000 (0.2651)  cls_action: 7.8649 (7.8434)  past_cls_action: 5.2544 (5.4451)  feat: 1.0473 (1.0627)  time: 2.4883  data: 1.1657  max mem: 3444
[2022-02-10 18:14:51,659][func.train][INFO] - [] Test:  [1510/1657]  eta: 0:06:58  iter_time: 2.5446  data_time: 1.2288  loss: 14.1906 (14.3496)  acc1/action: 0.0000 (0.0662)  acc5/action: 0.0000 (0.2647)  cls_action: 7.8649 (7.8433)  past_cls_action: 5.2544 (5.4436)  feat: 1.0549 (1.0627)  time: 2.5446  data: 1.2288  max mem: 3444
[2022-02-10 18:14:57,113][func.train][INFO] - [] Test:  [1512/1657]  eta: 0:06:52  iter_time: 2.5592  data_time: 1.2434  loss: 13.7177 (14.3483)  acc1/action: 0.0000 (0.0661)  acc5/action: 0.0000 (0.2644)  cls_action: 7.8667 (7.8434)  past_cls_action: 4.9315 (5.4422)  feat: 1.0563 (1.0627)  time: 2.5592  data: 1.2434  max mem: 3444
[2022-02-10 18:15:02,661][func.train][INFO] - [] Test:  [1514/1657]  eta: 0:06:47  iter_time: 2.5858  data_time: 1.2629  loss: 13.7177 (14.3502)  acc1/action: 0.0000 (0.0660)  acc5/action: 0.0000 (0.2640)  cls_action: 7.8667 (7.8434)  past_cls_action: 4.9315 (5.4442)  feat: 1.0596 (1.0627)  time: 2.5858  data: 1.2629  max mem: 3444
[2022-02-10 18:15:07,977][func.train][INFO] - [] Test:  [1516/1657]  eta: 0:06:41  iter_time: 2.6017  data_time: 1.2829  loss: 13.5636 (14.3485)  acc1/action: 0.0000 (0.0659)  acc5/action: 0.0000 (0.2637)  cls_action: 7.8642 (7.8435)  past_cls_action: 4.6365 (5.4424)  feat: 1.0616 (1.0627)  time: 2.6017  data: 1.2829  max mem: 3444
[2022-02-10 18:15:13,576][func.train][INFO] - [] Test:  [1518/1657]  eta: 0:06:35  iter_time: 2.6465  data_time: 1.3216  loss: 13.7177 (14.3512)  acc1/action: 0.0000 (0.0658)  acc5/action: 0.0000 (0.2633)  cls_action: 7.8667 (7.8436)  past_cls_action: 4.9315 (5.4449)  feat: 1.0682 (1.0627)  time: 2.6465  data: 1.3216  max mem: 3444
[2022-02-10 18:15:19,238][func.train][INFO] - [] Test:  [1520/1657]  eta: 0:06:30  iter_time: 2.6778  data_time: 1.3512  loss: 13.5636 (14.3506)  acc1/action: 0.0000 (0.0657)  acc5/action: 0.0000 (0.2630)  cls_action: 7.8725 (7.8437)  past_cls_action: 4.6365 (5.4442)  feat: 1.0668 (1.0627)  time: 2.6778  data: 1.3512  max mem: 3444
[2022-02-10 18:15:24,672][func.train][INFO] - [] Test:  [1522/1657]  eta: 0:06:24  iter_time: 2.6941  data_time: 1.3673  loss: 13.6523 (14.3502)  acc1/action: 0.0000 (0.0657)  acc5/action: 0.0000 (0.2626)  cls_action: 7.8694 (7.8437)  past_cls_action: 4.7013 (5.4438)  feat: 1.0682 (1.0627)  time: 2.6941  data: 1.3673  max mem: 3444
[2022-02-10 18:15:30,010][func.train][INFO] - [] Test:  [1524/1657]  eta: 0:06:18  iter_time: 2.6879  data_time: 1.3687  loss: 13.6523 (14.3494)  acc1/action: 0.0000 (0.0656)  acc5/action: 0.0000 (0.2623)  cls_action: 7.8694 (7.8438)  past_cls_action: 4.7013 (5.4428)  feat: 1.0668 (1.0628)  time: 2.6879  data: 1.3687  max mem: 3444
[2022-02-10 18:15:35,508][func.train][INFO] - [] Test:  [1526/1657]  eta: 0:06:12  iter_time: 2.7065  data_time: 1.3818  loss: 13.7177 (14.3521)  acc1/action: 0.0000 (0.0655)  acc5/action: 0.0000 (0.2620)  cls_action: 7.8728 (7.8438)  past_cls_action: 4.9315 (5.4454)  feat: 1.0699 (1.0628)  time: 2.7065  data: 1.3818  max mem: 3444
[2022-02-10 18:15:41,121][func.train][INFO] - [] Test:  [1528/1657]  eta: 0:06:07  iter_time: 2.7453  data_time: 1.4205  loss: 14.3523 (14.3536)  acc1/action: 0.0000 (0.0654)  acc5/action: 0.0000 (0.2616)  cls_action: 7.8728 (7.8439)  past_cls_action: 5.4585 (5.4468)  feat: 1.0726 (1.0629)  time: 2.7453  data: 1.4205  max mem: 3444
[2022-02-10 18:15:46,505][func.train][INFO] - [] Test:  [1530/1657]  eta: 0:06:01  iter_time: 2.7416  data_time: 1.4190  loss: 14.3523 (14.3505)  acc1/action: 0.0000 (0.0653)  acc5/action: 0.0000 (0.2830)  cls_action: 7.8728 (7.8438)  past_cls_action: 5.4585 (5.4438)  feat: 1.0713 (1.0629)  time: 2.7416  data: 1.4190  max mem: 3444
[2022-02-10 18:15:52,044][func.train][INFO] - [] Test:  [1532/1657]  eta: 0:05:55  iter_time: 2.7459  data_time: 1.4228  loss: 14.7834 (14.3506)  acc1/action: 0.0000 (0.0652)  acc5/action: 0.0000 (0.2827)  cls_action: 7.8704 (7.8439)  past_cls_action: 5.7440 (5.4438)  feat: 1.0713 (1.0629)  time: 2.7459  data: 1.4228  max mem: 3444
[2022-02-10 18:15:56,608][func.train][INFO] - [] Test:  [1534/1657]  eta: 0:05:49  iter_time: 2.6967  data_time: 1.3816  loss: 14.3523 (14.3489)  acc1/action: 0.0000 (0.0651)  acc5/action: 0.0000 (0.2823)  cls_action: 7.8728 (7.8439)  past_cls_action: 5.4585 (5.4422)  feat: 1.0703 (1.0628)  time: 2.6967  data: 1.3816  max mem: 3444
[2022-02-10 18:16:01,676][func.train][INFO] - [] Test:  [1536/1657]  eta: 0:05:44  iter_time: 2.6843  data_time: 1.3650  loss: 14.3523 (14.3460)  acc1/action: 0.0000 (0.0651)  acc5/action: 0.0000 (0.2819)  cls_action: 7.8728 (7.8439)  past_cls_action: 5.4585 (5.4394)  feat: 1.0699 (1.0628)  time: 2.6843  data: 1.3650  max mem: 3444
[2022-02-10 18:16:07,090][func.train][INFO] - [] Test:  [1538/1657]  eta: 0:05:38  iter_time: 2.6750  data_time: 1.3573  loss: 13.6152 (14.3440)  acc1/action: 0.0000 (0.0650)  acc5/action: 0.0000 (0.2816)  cls_action: 7.8585 (7.8438)  past_cls_action: 4.6950 (5.4374)  feat: 1.0688 (1.0628)  time: 2.6750  data: 1.3573  max mem: 3444
[2022-02-10 18:16:12,496][func.train][INFO] - [] Test:  [1540/1657]  eta: 0:05:32  iter_time: 2.6622  data_time: 1.3449  loss: 13.6523 (14.3441)  acc1/action: 0.0000 (0.0649)  acc5/action: 0.0000 (0.2812)  cls_action: 7.8563 (7.8438)  past_cls_action: 4.7013 (5.4376)  feat: 1.0688 (1.0627)  time: 2.6622  data: 1.3449  max mem: 3444
[2022-02-10 18:16:17,860][func.train][INFO] - [] Test:  [1542/1657]  eta: 0:05:27  iter_time: 2.6587  data_time: 1.3424  loss: 13.9810 (14.3452)  acc1/action: 0.0000 (0.0648)  acc5/action: 0.0000 (0.2808)  cls_action: 7.8563 (7.8439)  past_cls_action: 5.2019 (5.4387)  feat: 1.0655 (1.0627)  time: 2.6587  data: 1.3424  max mem: 3444
[2022-02-10 18:16:23,110][func.train][INFO] - [] Test:  [1544/1657]  eta: 0:05:21  iter_time: 2.6543  data_time: 1.3384  loss: 13.9810 (14.3446)  acc1/action: 0.0000 (0.0647)  acc5/action: 0.0000 (0.2805)  cls_action: 7.8463 (7.8439)  past_cls_action: 5.2019 (5.4380)  feat: 1.0655 (1.0627)  time: 2.6543  data: 1.3384  max mem: 3444
[2022-02-10 18:16:28,150][func.train][INFO] - [] Test:  [1546/1657]  eta: 0:05:15  iter_time: 2.6314  data_time: 1.3205  loss: 13.0587 (14.3424)  acc1/action: 0.0000 (0.0646)  acc5/action: 0.0000 (0.2801)  cls_action: 7.8344 (7.8438)  past_cls_action: 4.2071 (5.4359)  feat: 1.0655 (1.0627)  time: 2.6314  data: 1.3205  max mem: 3444
[2022-02-10 18:16:33,142][func.train][INFO] - [] Test:  [1548/1657]  eta: 0:05:09  iter_time: 2.6004  data_time: 1.2907  loss: 13.0076 (14.3393)  acc1/action: 0.0000 (0.0646)  acc5/action: 0.0000 (0.2798)  cls_action: 7.8327 (7.8437)  past_cls_action: 4.1594 (5.4329)  feat: 1.0520 (1.0627)  time: 2.6004  data: 1.2907  max mem: 3444
[2022-02-10 18:16:38,864][func.train][INFO] - [] Test:  [1550/1657]  eta: 0:05:04  iter_time: 2.6172  data_time: 1.2942  loss: 12.9840 (14.3360)  acc1/action: 0.0000 (0.0645)  acc5/action: 0.0000 (0.2794)  cls_action: 7.8327 (7.8436)  past_cls_action: 4.1493 (5.4297)  feat: 1.0502 (1.0627)  time: 2.6172  data: 1.2942  max mem: 3444
[2022-02-10 18:16:44,107][func.train][INFO] - [] Test:  [1552/1657]  eta: 0:04:58  iter_time: 2.6025  data_time: 1.2865  loss: 12.5979 (14.3314)  acc1/action: 0.0000 (0.0644)  acc5/action: 0.0000 (0.2790)  cls_action: 7.8319 (7.8436)  past_cls_action: 3.8726 (5.4252)  feat: 1.0356 (1.0626)  time: 2.6025  data: 1.2865  max mem: 3444
[2022-02-10 18:16:49,750][func.train][INFO] - [] Test:  [1554/1657]  eta: 0:04:52  iter_time: 2.6564  data_time: 1.3322  loss: 12.6162 (14.3292)  acc1/action: 0.0000 (0.0643)  acc5/action: 0.0000 (0.2787)  cls_action: 7.8231 (7.8434)  past_cls_action: 3.8792 (5.4232)  feat: 1.0402 (1.0626)  time: 2.6564  data: 1.3322  max mem: 3444
[2022-02-10 18:16:54,917][func.train][INFO] - [] Test:  [1556/1657]  eta: 0:04:47  iter_time: 2.6614  data_time: 1.3371  loss: 12.7256 (14.3263)  acc1/action: 0.0000 (0.0642)  acc5/action: 0.0000 (0.2783)  cls_action: 7.8284 (7.8434)  past_cls_action: 3.8953 (5.4203)  feat: 1.0402 (1.0626)  time: 2.6614  data: 1.3371  max mem: 3444
[2022-02-10 18:16:59,896][func.train][INFO] - [] Test:  [1558/1657]  eta: 0:04:41  iter_time: 2.6396  data_time: 1.3129  loss: 12.7256 (14.3239)  acc1/action: 0.0000 (0.0641)  acc5/action: 0.0000 (0.2780)  cls_action: 7.8089 (7.8434)  past_cls_action: 3.8953 (5.4180)  feat: 1.0356 (1.0625)  time: 2.6396  data: 1.3129  max mem: 3444
[2022-02-10 18:17:05,195][func.train][INFO] - [] Test:  [1560/1657]  eta: 0:04:35  iter_time: 2.6342  data_time: 1.3140  loss: 12.7256 (14.3234)  acc1/action: 0.0000 (0.0641)  acc5/action: 0.0000 (0.2776)  cls_action: 7.8129 (7.8434)  past_cls_action: 3.8953 (5.4176)  feat: 1.0402 (1.0625)  time: 2.6342  data: 1.3140  max mem: 3444
[2022-02-10 18:17:10,770][func.train][INFO] - [] Test:  [1562/1657]  eta: 0:04:29  iter_time: 2.6448  data_time: 1.3219  loss: 12.7256 (14.3255)  acc1/action: 0.0000 (0.0640)  acc5/action: 0.0000 (0.2772)  cls_action: 7.8062 (7.8432)  past_cls_action: 3.8953 (5.4197)  feat: 1.0443 (1.0625)  time: 2.6448  data: 1.3219  max mem: 3444
[2022-02-10 18:17:16,134][func.train][INFO] - [] Test:  [1564/1657]  eta: 0:04:24  iter_time: 2.6505  data_time: 1.3190  loss: 12.7256 (14.3255)  acc1/action: 0.0000 (0.0639)  acc5/action: 0.0000 (0.2769)  cls_action: 7.7939 (7.8432)  past_cls_action: 3.9113 (5.4198)  feat: 1.0458 (1.0625)  time: 2.6505  data: 1.3190  max mem: 3444
[2022-02-10 18:17:21,588][func.train][INFO] - [] Test:  [1566/1657]  eta: 0:04:18  iter_time: 2.6712  data_time: 1.3327  loss: 12.7256 (14.3235)  acc1/action: 0.0000 (0.0638)  acc5/action: 0.0000 (0.2765)  cls_action: 7.7939 (7.8432)  past_cls_action: 3.9113 (5.4179)  feat: 1.0458 (1.0625)  time: 2.6712  data: 1.3327  max mem: 3444
[2022-02-10 18:17:27,396][func.train][INFO] - [] Test:  [1568/1657]  eta: 0:04:12  iter_time: 2.7120  data_time: 1.3799  loss: 13.0196 (14.3231)  acc1/action: 0.0000 (0.0637)  acc5/action: 0.0000 (0.2762)  cls_action: 7.8057 (7.8432)  past_cls_action: 4.1382 (5.4175)  feat: 1.0473 (1.0625)  time: 2.7120  data: 1.3799  max mem: 3444
[2022-02-10 18:17:32,802][func.train][INFO] - [] Test:  [1570/1657]  eta: 0:04:07  iter_time: 2.6962  data_time: 1.3763  loss: 13.3574 (14.3230)  acc1/action: 0.0000 (0.0637)  acc5/action: 0.0000 (0.2758)  cls_action: 7.8066 (7.8432)  past_cls_action: 4.4678 (5.4174)  feat: 1.0481 (1.0625)  time: 2.6962  data: 1.3763  max mem: 3444
[2022-02-10 18:17:38,211][func.train][INFO] - [] Test:  [1572/1657]  eta: 0:04:01  iter_time: 2.7044  data_time: 1.3808  loss: 13.7270 (14.3233)  acc1/action: 0.0000 (0.0636)  acc5/action: 0.0000 (0.2755)  cls_action: 7.8066 (7.8432)  past_cls_action: 4.9185 (5.4176)  feat: 1.0503 (1.0625)  time: 2.7044  data: 1.3808  max mem: 3444
[2022-02-10 18:17:43,620][func.train][INFO] - [] Test:  [1574/1657]  eta: 0:03:55  iter_time: 2.6928  data_time: 1.3701  loss: 13.7986 (14.3247)  acc1/action: 0.0000 (0.0635)  acc5/action: 0.0000 (0.2751)  cls_action: 7.8066 (7.8431)  past_cls_action: 4.9666 (5.4192)  feat: 1.0573 (1.0625)  time: 2.6928  data: 1.3701  max mem: 3444
[2022-02-10 18:17:48,915][func.train][INFO] - [] Test:  [1576/1657]  eta: 0:03:50  iter_time: 2.6991  data_time: 1.3830  loss: 14.3435 (14.3250)  acc1/action: 0.0000 (0.0634)  acc5/action: 0.0000 (0.2748)  cls_action: 7.8066 (7.8431)  past_cls_action: 5.4680 (5.4194)  feat: 1.0591 (1.0625)  time: 2.6991  data: 1.3830  max mem: 3444
[2022-02-10 18:17:54,347][func.train][INFO] - [] Test:  [1578/1657]  eta: 0:03:44  iter_time: 2.7218  data_time: 1.4126  loss: 14.3557 (14.3245)  acc1/action: 0.0000 (0.0633)  acc5/action: 0.0000 (0.2744)  cls_action: 7.8066 (7.8431)  past_cls_action: 5.4870 (5.4190)  feat: 1.0591 (1.0625)  time: 2.7218  data: 1.4126  max mem: 3444
[2022-02-10 18:17:59,775][func.train][INFO] - [] Test:  [1580/1657]  eta: 0:03:38  iter_time: 2.7283  data_time: 1.4120  loss: 14.3596 (14.3256)  acc1/action: 0.0000 (0.0633)  acc5/action: 0.0000 (0.2741)  cls_action: 7.8034 (7.8430)  past_cls_action: 5.5179 (5.4202)  feat: 1.0591 (1.0625)  time: 2.7283  data: 1.4120  max mem: 3444
[2022-02-10 18:18:05,019][func.train][INFO] - [] Test:  [1582/1657]  eta: 0:03:32  iter_time: 2.7118  data_time: 1.3947  loss: 14.3596 (14.3283)  acc1/action: 0.0000 (0.0632)  acc5/action: 0.0000 (0.2737)  cls_action: 7.8066 (7.8431)  past_cls_action: 5.5179 (5.4228)  feat: 1.0591 (1.0625)  time: 2.7118  data: 1.3947  max mem: 3444
[2022-02-10 18:18:10,453][func.train][INFO] - [] Test:  [1584/1657]  eta: 0:03:27  iter_time: 2.7153  data_time: 1.4068  loss: 14.5990 (14.3294)  acc1/action: 0.0000 (0.0631)  acc5/action: 0.0000 (0.2734)  cls_action: 7.8066 (7.8431)  past_cls_action: 5.7307 (5.4239)  feat: 1.0597 (1.0625)  time: 2.7153  data: 1.4068  max mem: 3444
[2022-02-10 18:18:16,218][func.train][INFO] - [] Test:  [1586/1657]  eta: 0:03:21  iter_time: 2.7308  data_time: 1.4214  loss: 14.6186 (14.3291)  acc1/action: 0.0000 (0.0630)  acc5/action: 0.0000 (0.2731)  cls_action: 7.8066 (7.8431)  past_cls_action: 5.7660 (5.4235)  feat: 1.0599 (1.0625)  time: 2.7308  data: 1.4214  max mem: 3444
[2022-02-10 18:18:22,035][func.train][INFO] - [] Test:  [1588/1657]  eta: 0:03:15  iter_time: 2.7312  data_time: 1.4144  loss: 14.8079 (14.3298)  acc1/action: 0.0000 (0.0629)  acc5/action: 0.0000 (0.2727)  cls_action: 7.8105 (7.8431)  past_cls_action: 5.9577 (5.4242)  feat: 1.0599 (1.0625)  time: 2.7312  data: 1.4144  max mem: 3444
[2022-02-10 18:18:27,737][func.train][INFO] - [] Test:  [1590/1657]  eta: 0:03:10  iter_time: 2.7460  data_time: 1.4225  loss: 14.9130 (14.3320)  acc1/action: 0.0000 (0.0629)  acc5/action: 0.0000 (0.2724)  cls_action: 7.8105 (7.8431)  past_cls_action: 6.0036 (5.4264)  feat: 1.0599 (1.0625)  time: 2.7460  data: 1.4225  max mem: 3444
[2022-02-10 18:18:33,435][func.train][INFO] - [] Test:  [1592/1657]  eta: 0:03:04  iter_time: 2.7605  data_time: 1.4417  loss: 15.1331 (14.3346)  acc1/action: 0.0000 (0.0628)  acc5/action: 0.0000 (0.2720)  cls_action: 7.8430 (7.8432)  past_cls_action: 6.2204 (5.4288)  feat: 1.0626 (1.0625)  time: 2.7605  data: 1.4417  max mem: 3444
[2022-02-10 18:18:38,850][func.train][INFO] - [] Test:  [1594/1657]  eta: 0:02:58  iter_time: 2.7608  data_time: 1.4426  loss: 15.1331 (14.3368)  acc1/action: 0.0000 (0.0627)  acc5/action: 0.0000 (0.2717)  cls_action: 7.8514 (7.8433)  past_cls_action: 6.2204 (5.4309)  feat: 1.0689 (1.0626)  time: 2.7608  data: 1.4426  max mem: 3444
[2022-02-10 18:18:44,032][func.train][INFO] - [] Test:  [1596/1657]  eta: 0:02:53  iter_time: 2.7552  data_time: 1.4368  loss: 15.1331 (14.3309)  acc1/action: 0.0000 (0.0626)  acc5/action: 0.0000 (0.2922)  cls_action: 7.8506 (7.8432)  past_cls_action: 6.2204 (5.4251)  feat: 1.0740 (1.0626)  time: 2.7552  data: 1.4368  max mem: 3444
[2022-02-10 18:18:49,860][func.train][INFO] - [] Test:  [1598/1657]  eta: 0:02:47  iter_time: 2.7750  data_time: 1.4514  loss: 15.3615 (14.3315)  acc1/action: 0.0000 (0.0625)  acc5/action: 0.0000 (0.2918)  cls_action: 7.8514 (7.8432)  past_cls_action: 6.3374 (5.4257)  feat: 1.0740 (1.0626)  time: 2.7750  data: 1.4514  max mem: 3444
[2022-02-10 18:18:55,280][func.train][INFO] - [] Test:  [1600/1657]  eta: 0:02:41  iter_time: 2.7745  data_time: 1.4522  loss: 15.4300 (14.3343)  acc1/action: 0.0000 (0.0625)  acc5/action: 0.0000 (0.2915)  cls_action: 7.8514 (7.8431)  past_cls_action: 6.5447 (5.4286)  feat: 1.0740 (1.0626)  time: 2.7745  data: 1.4522  max mem: 3444
[2022-02-10 18:19:00,676][func.train][INFO] - [] Test:  [1602/1657]  eta: 0:02:36  iter_time: 2.7822  data_time: 1.4670  loss: 15.1331 (14.3332)  acc1/action: 0.0000 (0.0624)  acc5/action: 0.0000 (0.2911)  cls_action: 7.8467 (7.8430)  past_cls_action: 6.2204 (5.4275)  feat: 1.0735 (1.0626)  time: 2.7822  data: 1.4670  max mem: 3444
[2022-02-10 18:19:05,999][func.train][INFO] - [] Test:  [1604/1657]  eta: 0:02:30  iter_time: 2.7766  data_time: 1.4553  loss: 15.1331 (14.3338)  acc1/action: 0.0000 (0.0623)  acc5/action: 0.0000 (0.2908)  cls_action: 7.8506 (7.8431)  past_cls_action: 6.2204 (5.4281)  feat: 1.0735 (1.0626)  time: 2.7766  data: 1.4553  max mem: 3444
[2022-02-10 18:19:10,299][func.train][INFO] - [] Test:  [1606/1657]  eta: 0:02:24  iter_time: 2.7034  data_time: 1.3778  loss: 15.1331 (14.3302)  acc1/action: 0.0000 (0.0622)  acc5/action: 0.0000 (0.2904)  cls_action: 7.8514 (7.8432)  past_cls_action: 6.2204 (5.4245)  feat: 1.0732 (1.0625)  time: 2.7034  data: 1.3778  max mem: 3444
[2022-02-10 18:19:15,561][func.train][INFO] - [] Test:  [1608/1657]  eta: 0:02:19  iter_time: 2.6757  data_time: 1.3583  loss: 14.4148 (14.3283)  acc1/action: 0.0000 (0.0622)  acc5/action: 0.0000 (0.2900)  cls_action: 7.8554 (7.8432)  past_cls_action: 5.4860 (5.4226)  feat: 1.0714 (1.0625)  time: 2.6757  data: 1.3583  max mem: 3444
[2022-02-10 18:19:21,184][func.train][INFO] - [] Test:  [1610/1657]  eta: 0:02:13  iter_time: 2.6717  data_time: 1.3501  loss: 13.6351 (14.3246)  acc1/action: 0.0000 (0.0621)  acc5/action: 0.0000 (0.2897)  cls_action: 7.8602 (7.8433)  past_cls_action: 4.7086 (5.4188)  feat: 1.0699 (1.0625)  time: 2.6717  data: 1.3501  max mem: 3444
[2022-02-10 18:19:26,400][func.train][INFO] - [] Test:  [1612/1657]  eta: 0:02:07  iter_time: 2.6476  data_time: 1.3253  loss: 13.6303 (14.3235)  acc1/action: 0.0000 (0.0620)  acc5/action: 0.0000 (0.2893)  cls_action: 7.8602 (7.8435)  past_cls_action: 4.6986 (5.4176)  feat: 1.0621 (1.0625)  time: 2.6476  data: 1.3253  max mem: 3444
[2022-02-10 18:19:31,792][func.train][INFO] - [] Test:  [1614/1657]  eta: 0:02:02  iter_time: 2.6465  data_time: 1.3313  loss: 13.2223 (14.3234)  acc1/action: 0.0000 (0.0619)  acc5/action: 0.0000 (0.2890)  cls_action: 7.8554 (7.8435)  past_cls_action: 4.2067 (5.4174)  feat: 1.0621 (1.0625)  time: 2.6465  data: 1.3313  max mem: 3444
[2022-02-10 18:19:37,039][func.train][INFO] - [] Test:  [1616/1657]  eta: 0:01:56  iter_time: 2.6497  data_time: 1.3252  loss: 13.2223 (14.3190)  acc1/action: 0.0000 (0.0618)  acc5/action: 0.0000 (0.2886)  cls_action: 7.8727 (7.8436)  past_cls_action: 4.2067 (5.4130)  feat: 1.0573 (1.0624)  time: 2.6497  data: 1.3252  max mem: 3444
[2022-02-10 18:19:42,321][func.train][INFO] - [] Test:  [1618/1657]  eta: 0:01:50  iter_time: 2.6224  data_time: 1.3016  loss: 13.2223 (14.3201)  acc1/action: 0.0000 (0.0618)  acc5/action: 0.0000 (0.2882)  cls_action: 7.8882 (7.8437)  past_cls_action: 4.2067 (5.4139)  feat: 1.0573 (1.0625)  time: 2.6224  data: 1.3016  max mem: 3444
[2022-02-10 18:19:47,762][func.train][INFO] - [] Test:  [1620/1657]  eta: 0:01:44  iter_time: 2.6234  data_time: 1.3011  loss: 13.0056 (14.3180)  acc1/action: 0.0000 (0.0617)  acc5/action: 0.0000 (0.2879)  cls_action: 7.8882 (7.8436)  past_cls_action: 4.1194 (5.4119)  feat: 1.0501 (1.0625)  time: 2.6234  data: 1.3011  max mem: 3444
[2022-02-10 18:19:53,228][func.train][INFO] - [] Test:  [1622/1657]  eta: 0:01:39  iter_time: 2.6269  data_time: 1.3065  loss: 13.2223 (14.3184)  acc1/action: 0.0000 (0.0616)  acc5/action: 0.0000 (0.2875)  cls_action: 7.9084 (7.8437)  past_cls_action: 4.2067 (5.4121)  feat: 1.0573 (1.0625)  time: 2.6269  data: 1.3065  max mem: 3444
[2022-02-10 18:19:58,518][func.train][INFO] - [] Test:  [1624/1657]  eta: 0:01:33  iter_time: 2.6253  data_time: 1.3018  loss: 12.5549 (14.3173)  acc1/action: 0.0000 (0.0615)  acc5/action: 0.0000 (0.2872)  cls_action: 7.9203 (7.8438)  past_cls_action: 3.6624 (5.4109)  feat: 1.0501 (1.0625)  time: 2.6253  data: 1.3018  max mem: 3444
[2022-02-10 18:20:03,950][func.train][INFO] - [] Test:  [1626/1657]  eta: 0:01:27  iter_time: 2.6819  data_time: 1.3637  loss: 12.5549 (14.3143)  acc1/action: 0.0000 (0.0615)  acc5/action: 0.0000 (0.2868)  cls_action: 7.9143 (7.8438)  past_cls_action: 3.6624 (5.4080)  feat: 1.0573 (1.0625)  time: 2.6819  data: 1.3637  max mem: 3444
[2022-02-10 18:20:09,453][func.train][INFO] - [] Test:  [1628/1657]  eta: 0:01:22  iter_time: 2.6939  data_time: 1.3680  loss: 12.5549 (14.3101)  acc1/action: 0.0000 (0.0614)  acc5/action: 0.0000 (0.2865)  cls_action: 7.9203 (7.8438)  past_cls_action: 3.6117 (5.4037)  feat: 1.0573 (1.0625)  time: 2.6939  data: 1.3680  max mem: 3444
[2022-02-10 18:20:14,782][func.train][INFO] - [] Test:  [1630/1657]  eta: 0:01:16  iter_time: 2.6792  data_time: 1.3660  loss: 12.6305 (14.3066)  acc1/action: 0.0000 (0.0613)  acc5/action: 0.0000 (0.2861)  cls_action: 7.9203 (7.8440)  past_cls_action: 3.6624 (5.4002)  feat: 1.0615 (1.0625)  time: 2.6792  data: 1.3660  max mem: 3444
[2022-02-10 18:20:20,098][func.train][INFO] - [] Test:  [1632/1657]  eta: 0:01:10  iter_time: 2.6842  data_time: 1.3746  loss: 12.5549 (14.3043)  acc1/action: 0.0000 (0.0612)  acc5/action: 0.0000 (0.2858)  cls_action: 7.9076 (7.8440)  past_cls_action: 3.6117 (5.3978)  feat: 1.0655 (1.0625)  time: 2.6842  data: 1.3746  max mem: 3444
[2022-02-10 18:20:25,657][func.train][INFO] - [] Test:  [1634/1657]  eta: 0:01:05  iter_time: 2.6925  data_time: 1.3712  loss: 12.1726 (14.3024)  acc1/action: 0.0000 (0.0612)  acc5/action: 0.0000 (0.2854)  cls_action: 7.9076 (7.8441)  past_cls_action: 3.4003 (5.3959)  feat: 1.0690 (1.0625)  time: 2.6925  data: 1.3712  max mem: 3444
[2022-02-10 18:20:30,972][func.train][INFO] - [] Test:  [1636/1657]  eta: 0:00:59  iter_time: 2.6959  data_time: 1.3817  loss: 12.6305 (14.3008)  acc1/action: 0.0000 (0.0611)  acc5/action: 0.0000 (0.2851)  cls_action: 7.9076 (7.8442)  past_cls_action: 3.6117 (5.3942)  feat: 1.0690 (1.0624)  time: 2.6959  data: 1.3817  max mem: 3444
[2022-02-10 18:20:36,353][func.train][INFO] - [] Test:  [1638/1657]  eta: 0:00:53  iter_time: 2.7008  data_time: 1.3881  loss: 12.1726 (14.2983)  acc1/action: 0.0000 (0.0610)  acc5/action: 0.0000 (0.2847)  cls_action: 7.9133 (7.8443)  past_cls_action: 3.4003 (5.3916)  feat: 1.0554 (1.0624)  time: 2.7008  data: 1.3881  max mem: 3444
[2022-02-10 18:20:41,914][func.train][INFO] - [] Test:  [1640/1657]  eta: 0:00:48  iter_time: 2.7069  data_time: 1.3950  loss: 12.1032 (14.2954)  acc1/action: 0.0000 (0.0609)  acc5/action: 0.0000 (0.2844)  cls_action: 7.9133 (7.8443)  past_cls_action: 3.1629 (5.3887)  feat: 1.0554 (1.0624)  time: 2.7069  data: 1.3950  max mem: 3444
[2022-02-10 18:20:47,352][func.train][INFO] - [] Test:  [1642/1657]  eta: 0:00:42  iter_time: 2.7054  data_time: 1.3851  loss: 12.1032 (14.2946)  acc1/action: 0.0000 (0.0609)  acc5/action: 0.0000 (0.2840)  cls_action: 7.9076 (7.8444)  past_cls_action: 3.1629 (5.3879)  feat: 1.0538 (1.0624)  time: 2.7054  data: 1.3851  max mem: 3444
[2022-02-10 18:20:52,928][func.train][INFO] - [] Test:  [1644/1657]  eta: 0:00:36  iter_time: 2.7198  data_time: 1.4068  loss: 12.1032 (14.2925)  acc1/action: 0.0000 (0.0608)  acc5/action: 0.0000 (0.2837)  cls_action: 7.8942 (7.8444)  past_cls_action: 3.1629 (5.3858)  feat: 1.0541 (1.0624)  time: 2.7198  data: 1.4068  max mem: 3444
[2022-02-10 18:20:58,342][func.train][INFO] - [] Test:  [1646/1657]  eta: 0:00:31  iter_time: 2.7188  data_time: 1.4125  loss: 12.1726 (14.2916)  acc1/action: 0.0000 (0.0607)  acc5/action: 0.0000 (0.2833)  cls_action: 7.8942 (7.8444)  past_cls_action: 3.4003 (5.3848)  feat: 1.0541 (1.0624)  time: 2.7188  data: 1.4125  max mem: 3444
[2022-02-10 18:21:03,830][func.train][INFO] - [] Test:  [1648/1657]  eta: 0:00:25  iter_time: 2.7181  data_time: 1.4117  loss: 12.6964 (14.2900)  acc1/action: 0.0000 (0.0606)  acc5/action: 0.0000 (0.2830)  cls_action: 7.8843 (7.8443)  past_cls_action: 3.6848 (5.3833)  feat: 1.0554 (1.0625)  time: 2.7181  data: 1.4117  max mem: 3444
[2022-02-10 18:21:09,389][func.train][INFO] - [] Test:  [1650/1657]  eta: 0:00:19  iter_time: 2.7296  data_time: 1.4140  loss: 12.9138 (14.2915)  acc1/action: 0.0000 (0.0606)  acc5/action: 0.0000 (0.2827)  cls_action: 7.8814 (7.8443)  past_cls_action: 4.1274 (5.3846)  feat: 1.0703 (1.0625)  time: 2.7296  data: 1.4140  max mem: 3444
[2022-02-10 18:21:14,908][func.train][INFO] - [] Test:  [1652/1657]  eta: 0:00:14  iter_time: 2.7398  data_time: 1.4154  loss: 13.0872 (14.2936)  acc1/action: 0.0000 (0.0605)  acc5/action: 0.0000 (0.2823)  cls_action: 7.8670 (7.8443)  past_cls_action: 4.1881 (5.3868)  feat: 1.0730 (1.0625)  time: 2.7398  data: 1.4154  max mem: 3444
[2022-02-10 18:21:20,196][func.train][INFO] - [] Test:  [1654/1657]  eta: 0:00:08  iter_time: 2.7262  data_time: 1.4125  loss: 13.0872 (14.2913)  acc1/action: 0.0000 (0.0604)  acc5/action: 0.0000 (0.2820)  cls_action: 7.8566 (7.8443)  past_cls_action: 4.1881 (5.3844)  feat: 1.0762 (1.0625)  time: 2.7262  data: 1.4125  max mem: 3444
[2022-02-10 18:21:23,922][func.train][INFO] - [] Test:  [1656/1657]  eta: 0:00:02  iter_time: 2.6468  data_time: 1.3715  loss: 12.9551 (14.2888)  acc1/action: 0.0000 (0.0604)  acc5/action: 0.0000 (0.2816)  cls_action: 7.8566 (7.8445)  past_cls_action: 4.1274 (5.3818)  feat: 1.0762 (1.0625)  time: 2.6468  data: 1.3715  max mem: 3444
[2022-02-10 18:21:23,922][func.train][INFO] - [] Test: Total time: 1:18:15
[2022-02-10 18:21:27,655][root][INFO] - Reading from resfiles
[2022-02-10 18:21:34,000][func.train][INFO] - []
[2022-02-10 18:21:34,001][root][INFO] - iter_time: 2.833171
[2022-02-10 18:21:34,001][root][INFO] - data_time: 1.509094
[2022-02-10 18:21:34,001][root][INFO] - loss: 14.288804
[2022-02-10 18:21:34,001][root][INFO] - acc1/action: 0.060350
[2022-02-10 18:21:34,001][root][INFO] - acc5/action: 0.281633
[2022-02-10 18:21:34,001][root][INFO] - cls_action: 7.844472
[2022-02-10 18:21:34,001][root][INFO] - past_cls_action: 5.381803
[2022-02-10 18:21:34,002][root][INFO] - feat: 1.062530

[Anirudh257]

@CodyQ3 you have to do

from notebooks.utils import *
CFG_FILES = [
('expts/01_ek100_avt.txt', 0),
('expts/03_ek100_avt_tsn_obj.txt', 0),
]
WTS = [2.5, 0.5]
print_accuracies_epic(get_epic_marginalize_late_fuse(CFG_FILES, weights=WTS)[0])

This will print the entire result: acc1/acc5.

[rohitgirdhar]

you are running it in test_only mode but not initializing with the trained checkpoint (it's initializing from the VIT checkpoint, not the AVT trained checkpoint). You can set the checkpoint in the experiment config using something like this:

train.init_from_model=[[OUTPUTS/expts/01_ek100_avt.txt/0/checkpoint.pth]]

[CodyQ3]

@rohitgirdhar Thank you. Now, the final results seem much more in line with the reported ones:

2022-02-14 10:20:23,897][root][INFO] - iter_time: 2.586582
[2022-02-14 10:20:23,897][root][INFO] - data_time: 1.433165
[2022-02-14 10:20:23,897][root][INFO] - loss: 9.411732
[2022-02-14 10:20:23,897][root][INFO] - acc1/action: 12.311406
[2022-02-14 10:20:23,897][root][INFO] - acc5/action: 30.054314
[2022-02-14 10:20:23,898][root][INFO] - cls_action: 5.545769
[2022-02-14 10:20:23,898][root][INFO] - past_cls_action: 3.487312
[2022-02-14 10:20:23,898][root][INFO] - feat: 0.378651

What is the meaning of the metric-related values printed in each line of the log (e.g., [2022-02-14 10:20:13,526][func.train][INFO] - [] Test: [1656/1657] eta: 0:00:02 iter_time: 2.4252 data_time: 1.3088 loss: 9.6064 (9.4117) acc1/action: 0.0000 (12.3114) acc5/action: 0.0000 (30.0543) cls_action: 5.9764 (5.5458) past_cls_action: 2.7023 (3.4873) feat: 0.3410 (0.3787) time: 2.4252 data: 1.3088 max mem: 3439)?

For acc1/action, I thought the first value (0.0000) means the current accuracy of the iteration and the value in parenthesis (12.3114) the average accuracy so far. However, it seems not to be the case since acc1/action is still 0.0000 (xxx) most of the time. I have attached the entire log.
AVT.log

[rohitgirdhar]

Hi @CodyQ3 , ah yes you can ignore that part of the log; it should ideally print the running accuracy but I never got around to fixing it.

[CodyQ3]

How long does an inference take? Is the model real-time capable?