Add files using upload-large-folder tool

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Camera_Motion.log

@@ -0,0 +1,332 @@
+[2026-06-08 18:42:24 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Camera_Motion --dimension Camera_Motion --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Camera_Motion', dimension=['Camera_Motion'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion/results_2026-06-08-18:42:29_full_info.json
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion/results_2026-06-08-18:42:29_full_info.json
+
  0%|          | 0/108 [00:00<?, ?it/s]2026-06-08 18:42:52,609 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:43:00,374 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:43:07,209 - pyscenedetect - INFO - Detecting scenes...
+
  1%|          | 1/108 [00:21<38:20, 21.50s/it]2026-06-08 18:43:14,116 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:43:21,046 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:43:27,832 - pyscenedetect - INFO - Detecting scenes...
+
  2%|▏         | 2/108 [00:42<37:24, 21.17s/it]2026-06-08 18:43:35,047 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:43:42,092 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:43:49,268 - pyscenedetect - INFO - Detecting scenes...
+
  3%|▎         | 3/108 [01:03<37:15, 21.29s/it]2026-06-08 18:43:56,485 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:44:03,897 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:44:10,791 - pyscenedetect - INFO - Detecting scenes...
+
  4%|▎         | 4/108 [01:24<36:33, 21.09s/it]2026-06-08 18:44:17,264 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:44:24,167 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:44:31,054 - pyscenedetect - INFO - Detecting scenes...
+
  5%|▍         | 5/108 [01:45<36:10, 21.07s/it]2026-06-08 18:44:38,288 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:44:45,162 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:44:51,858 - pyscenedetect - INFO - Detecting scenes...
+
  6%|▌         | 6/108 [02:07<36:17, 21.35s/it]2026-06-08 18:45:00,207 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:45:07,578 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:45:14,822 - pyscenedetect - INFO - Detecting scenes...
+
  6%|▋         | 7/108 [02:29<36:01, 21.41s/it]2026-06-08 18:45:21,725 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:45:28,787 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:45:35,728 - pyscenedetect - INFO - Detecting scenes...
+
  7%|▋         | 8/108 [02:50<35:26, 21.26s/it]2026-06-08 18:45:42,694 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:45:49,849 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:45:57,172 - pyscenedetect - INFO - Detecting scenes...
+
  8%|▊         | 9/108 [03:11<35:08, 21.30s/it]2026-06-08 18:46:04,072 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:46:10,720 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:46:17,642 - pyscenedetect - INFO - Detecting scenes...
+
  9%|▉         | 10/108 [03:31<34:12, 20.94s/it]2026-06-08 18:46:24,174 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:46:31,017 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:46:37,867 - pyscenedetect - INFO - Detecting scenes...
+
 10%|█         | 11/108 [03:52<33:35, 20.78s/it]2026-06-08 18:46:44,618 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:46:51,416 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:46:58,346 - pyscenedetect - INFO - Detecting scenes...
+
 11%|█         | 12/108 [04:12<32:56, 20.59s/it]2026-06-08 18:47:04,754 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:47:11,633 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:47:17,918 - pyscenedetect - INFO - Detecting scenes...
+
 12%|█▏        | 13/108 [04:30<31:42, 20.03s/it]2026-06-08 18:47:23,479 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:47:28,505 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:47:33,852 - pyscenedetect - INFO - Detecting scenes...
+
 13%|█▎        | 14/108 [04:46<29:20, 18.73s/it]2026-06-08 18:47:39,223 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:47:44,530 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:47:49,248 - pyscenedetect - INFO - Detecting scenes...
+
 14%|█▍        | 15/108 [05:00<26:53, 17.35s/it]2026-06-08 18:47:53,345 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:47:57,581 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:01,795 - pyscenedetect - INFO - Detecting scenes...
+
 15%|█▍        | 16/108 [05:13<24:26, 15.94s/it]2026-06-08 18:48:06,006 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:10,256 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:14,712 - pyscenedetect - INFO - Detecting scenes...
+
 16%|█▌        | 17/108 [05:26<22:45, 15.00s/it]2026-06-08 18:48:18,839 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:22,998 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:27,334 - pyscenedetect - INFO - Detecting scenes...
+
 17%|█▋        | 18/108 [05:38<21:26, 14.29s/it]2026-06-08 18:48:31,481 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:35,624 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:39,731 - pyscenedetect - INFO - Detecting scenes...
+
 18%|█▊        | 19/108 [05:51<20:25, 13.77s/it]2026-06-08 18:48:44,034 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:48,213 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:48:52,340 - pyscenedetect - INFO - Detecting scenes...
+
 19%|█▊        | 20/108 [06:03<19:37, 13.38s/it]2026-06-08 18:48:56,495 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:00,720 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:04,880 - pyscenedetect - INFO - Detecting scenes...
+
 19%|█▉        | 21/108 [06:16<19:05, 13.17s/it]2026-06-08 18:49:09,168 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:13,312 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:17,496 - pyscenedetect - INFO - Detecting scenes...
+
 20%|██        | 22/108 [06:29<18:33, 12.95s/it]2026-06-08 18:49:21,617 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:25,831 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:29,920 - pyscenedetect - INFO - Detecting scenes...
+
 21%|██▏       | 23/108 [06:41<18:07, 12.79s/it]2026-06-08 18:49:34,028 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:38,227 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:42,278 - pyscenedetect - INFO - Detecting scenes...
+
 22%|██▏       | 24/108 [06:53<17:45, 12.69s/it]2026-06-08 18:49:46,472 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:50,715 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:49:55,091 - pyscenedetect - INFO - Detecting scenes...
+
 23%|██▎       | 25/108 [07:06<17:32, 12.68s/it]2026-06-08 18:49:59,152 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:03,320 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:07,530 - pyscenedetect - INFO - Detecting scenes...
+
 24%|██▍       | 26/108 [07:18<17:13, 12.61s/it]2026-06-08 18:50:11,579 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:15,674 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:20,732 - pyscenedetect - INFO - Detecting scenes...
+
 25%|██▌       | 27/108 [07:32<17:25, 12.91s/it]2026-06-08 18:50:25,183 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:29,243 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:33,367 - pyscenedetect - INFO - Detecting scenes...
+
 26%|██▌       | 28/108 [07:45<17:01, 12.77s/it]2026-06-08 18:50:37,651 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:41,986 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:46,100 - pyscenedetect - INFO - Detecting scenes...
+
 27%|██▋       | 29/108 [07:57<16:45, 12.73s/it]2026-06-08 18:50:50,274 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:54,810 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:50:59,186 - pyscenedetect - INFO - Detecting scenes...
+
 28%|██▊       | 30/108 [08:10<16:39, 12.81s/it]2026-06-08 18:51:03,268 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:07,448 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:11,750 - pyscenedetect - INFO - Detecting scenes...
+
 29%|██▊       | 31/108 [08:23<16:24, 12.78s/it]2026-06-08 18:51:15,984 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:20,119 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:24,500 - pyscenedetect - INFO - Detecting scenes...
+
 30%|██▉       | 32/108 [08:36<16:18, 12.87s/it]2026-06-08 18:51:29,069 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:33,109 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:37,227 - pyscenedetect - INFO - Detecting scenes...
+
 31%|███       | 33/108 [08:48<15:56, 12.75s/it]2026-06-08 18:51:41,533 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:45,822 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:49,907 - pyscenedetect - INFO - Detecting scenes...
+
 31%|███▏      | 34/108 [09:01<15:36, 12.65s/it]2026-06-08 18:51:53,953 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:51:58,376 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:02,796 - pyscenedetect - INFO - Detecting scenes...
+
 32%|███▏      | 35/108 [09:14<15:28, 12.72s/it]2026-06-08 18:52:06,824 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:10,922 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:15,247 - pyscenedetect - INFO - Detecting scenes...
+
 33%|███▎      | 36/108 [09:26<15:15, 12.71s/it]2026-06-08 18:52:19,513 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:23,661 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:27,908 - pyscenedetect - INFO - Detecting scenes...
+
 34%|███▍      | 37/108 [09:40<15:27, 13.07s/it]2026-06-08 18:52:33,427 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:38,437 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:43,474 - pyscenedetect - INFO - Detecting scenes...
+
 35%|███▌      | 38/108 [09:56<16:07, 13.82s/it]2026-06-08 18:52:48,977 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:52:54,363 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:53:01,177 - pyscenedetect - INFO - Detecting scenes...
+
 36%|███▌      | 39/108 [10:15<17:35, 15.29s/it]2026-06-08 18:53:07,725 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:53:14,359 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:53:21,993 - pyscenedetect - INFO - Detecting scenes...
+
 37%|███▋      | 40/108 [10:36<19:18, 17.04s/it]2026-06-08 18:53:28,836 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:53:35,323 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:53:41,987 - pyscenedetect - INFO - Detecting scenes...
+
 38%|███▊      | 41/108 [10:56<19:57, 17.87s/it]2026-06-08 18:53:48,667 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:53:55,456 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:54:02,519 - pyscenedetect - INFO - Detecting scenes...
+
 39%|███▉      | 42/108 [11:16<20:30, 18.65s/it]2026-06-08 18:54:09,129 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:54:15,648 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:54:21,953 - pyscenedetect - INFO - Detecting scenes...
+
 40%|███▉      | 43/108 [11:36<20:30, 18.94s/it]2026-06-08 18:54:28,733 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:54:35,285 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:54:42,441 - pyscenedetect - INFO - Detecting scenes...
+
 41%|████      | 44/108 [11:57<20:52, 19.56s/it]2026-06-08 18:54:49,750 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:54:56,397 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:55:02,783 - pyscenedetect - INFO - Detecting scenes...
+
 42%|████▏     | 45/108 [12:16<20:37, 19.64s/it]2026-06-08 18:55:09,573 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:55:16,513 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:55:23,650 - pyscenedetect - INFO - Detecting scenes...
+
 43%|████▎     | 46/108 [12:38<20:48, 20.14s/it]2026-06-08 18:55:30,881 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:55:37,375 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:55:44,631 - pyscenedetect - INFO - Detecting scenes...
+
 44%|████▎     | 47/108 [12:58<20:21, 20.03s/it]2026-06-08 18:55:50,682 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:55:57,524 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:56:04,429 - pyscenedetect - INFO - Detecting scenes...
+
 44%|████▍     | 48/108 [13:18<20:09, 20.15s/it]2026-06-08 18:56:11,157 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:56:17,438 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:56:23,979 - pyscenedetect - INFO - Detecting scenes...
+
 45%|████▌     | 49/108 [13:38<19:44, 20.08s/it]2026-06-08 18:56:31,012 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:56:37,847 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:56:44,638 - pyscenedetect - INFO - Detecting scenes...
+
 46%|████▋     | 50/108 [13:59<19:40, 20.35s/it]2026-06-08 18:56:52,007 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:56:59,097 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:57:06,171 - pyscenedetect - INFO - Detecting scenes...
+
 47%|████▋     | 51/108 [14:19<19:19, 20.34s/it]2026-06-08 18:57:12,314 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:57:18,803 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:57:25,472 - pyscenedetect - INFO - Detecting scenes...
+
 48%|████▊     | 52/108 [14:39<18:55, 20.27s/it]2026-06-08 18:57:32,434 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:57:39,594 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:57:46,695 - pyscenedetect - INFO - Detecting scenes...
+
 49%|████▉     | 53/108 [15:01<18:52, 20.60s/it]2026-06-08 18:57:53,792 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:58:00,588 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:58:07,451 - pyscenedetect - INFO - Detecting scenes...
+
 50%|█████     | 54/108 [15:21<18:31, 20.59s/it]2026-06-08 18:58:14,347 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:58:20,998 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:58:28,093 - pyscenedetect - INFO - Detecting scenes...
+
 51%|█████     | 55/108 [15:42<18:09, 20.56s/it]2026-06-08 18:58:34,863 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:58:41,242 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:58:48,028 - pyscenedetect - INFO - Detecting scenes...
+
 52%|█████▏    | 56/108 [16:02<17:40, 20.40s/it]2026-06-08 18:58:54,855 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:59:01,732 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:59:08,142 - pyscenedetect - INFO - Detecting scenes...
+
 53%|█████▎    | 57/108 [16:22<17:13, 20.26s/it]2026-06-08 18:59:14,819 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:59:22,007 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:59:29,219 - pyscenedetect - INFO - Detecting scenes...
+
 54%|█████▎    | 58/108 [16:43<17:02, 20.45s/it]2026-06-08 18:59:35,705 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:59:42,403 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:59:48,817 - pyscenedetect - INFO - Detecting scenes...
+
 55%|█████▍    | 59/108 [17:02<16:32, 20.25s/it]2026-06-08 18:59:55,480 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:02,151 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:07,905 - pyscenedetect - INFO - Detecting scenes...
+
 56%|█████▌    | 60/108 [17:20<15:27, 19.32s/it]2026-06-08 19:00:12,615 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:18,056 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:23,559 - pyscenedetect - INFO - Detecting scenes...
+
 56%|█████▋    | 61/108 [17:36<14:32, 18.56s/it]2026-06-08 19:00:29,408 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:34,292 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:39,597 - pyscenedetect - INFO - Detecting scenes...
+
 57%|█████▋    | 62/108 [17:51<13:14, 17.27s/it]2026-06-08 19:00:43,680 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:47,866 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:00:52,156 - pyscenedetect - INFO - Detecting scenes...
+
 58%|█████▊    | 63/108 [18:04<11:59, 15.99s/it]2026-06-08 19:00:56,672 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:01,213 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:05,995 - pyscenedetect - INFO - Detecting scenes...
+
 59%|█████▉    | 64/108 [18:17<11:09, 15.21s/it]2026-06-08 19:01:10,055 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:14,171 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:18,787 - pyscenedetect - INFO - Detecting scenes...
+
 60%|██████    | 65/108 [18:30<10:30, 14.66s/it]2026-06-08 19:01:23,423 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:27,447 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:31,486 - pyscenedetect - INFO - Detecting scenes...
+
 61%|██████    | 66/108 [18:43<09:45, 13.95s/it]2026-06-08 19:01:35,743 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:40,457 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:44,897 - pyscenedetect - INFO - Detecting scenes...
+
 62%|██████▏   | 67/108 [18:56<09:23, 13.75s/it]2026-06-08 19:01:49,012 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:53,154 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:01:57,812 - pyscenedetect - INFO - Detecting scenes...
+
 63%|██████▎   | 68/108 [19:09<09:05, 13.65s/it]2026-06-08 19:02:02,423 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:06,546 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:10,618 - pyscenedetect - INFO - Detecting scenes...
+
 64%|██████▍   | 69/108 [19:22<08:37, 13.26s/it]2026-06-08 19:02:14,796 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:19,211 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:23,359 - pyscenedetect - INFO - Detecting scenes...
+
 65%|██████▍   | 70/108 [19:34<08:17, 13.09s/it]2026-06-08 19:02:27,489 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:31,586 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:36,284 - pyscenedetect - INFO - Detecting scenes...
+
 66%|██████▌   | 71/108 [19:47<08:02, 13.05s/it]2026-06-08 19:02:40,426 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:44,468 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:48,790 - pyscenedetect - INFO - Detecting scenes...
+
 67%|██████▋   | 72/108 [20:00<07:44, 12.91s/it]2026-06-08 19:02:53,027 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:02:57,263 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:01,399 - pyscenedetect - INFO - Detecting scenes...
+
 68%|██████▊   | 73/108 [20:12<07:28, 12.81s/it]2026-06-08 19:03:05,594 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:09,855 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:14,033 - pyscenedetect - INFO - Detecting scenes...
+
 69%|██████▊   | 74/108 [20:25<07:14, 12.77s/it]2026-06-08 19:03:18,279 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:22,355 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:26,545 - pyscenedetect - INFO - Detecting scenes...
+
 69%|██████▉   | 75/108 [20:38<06:58, 12.67s/it]2026-06-08 19:03:30,720 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:34,926 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:39,150 - pyscenedetect - INFO - Detecting scenes...
+
 70%|███████   | 76/108 [20:50<06:44, 12.64s/it]2026-06-08 19:03:43,277 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:47,536 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:03:51,674 - pyscenedetect - INFO - Detecting scenes...
+
 71%|███████▏  | 77/108 [21:03<06:30, 12.59s/it]2026-06-08 19:03:55,740 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:00,037 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:04,330 - pyscenedetect - INFO - Detecting scenes...
+
 72%|███████▏  | 78/108 [21:15<06:18, 12.60s/it]2026-06-08 19:04:08,390 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:12,734 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:17,034 - pyscenedetect - INFO - Detecting scenes...
+
 73%|███████▎  | 79/108 [21:28<06:05, 12.61s/it]2026-06-08 19:04:21,014 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:25,070 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:29,345 - pyscenedetect - INFO - Detecting scenes...
+
 74%|███████▍  | 80/108 [21:41<05:56, 12.72s/it]2026-06-08 19:04:33,981 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:38,076 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:42,216 - pyscenedetect - INFO - Detecting scenes...
+
 75%|███████▌  | 81/108 [21:54<05:44, 12.77s/it]2026-06-08 19:04:46,874 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:51,493 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:04:55,550 - pyscenedetect - INFO - Detecting scenes...
+
 76%|███████▌  | 82/108 [22:07<05:32, 12.79s/it]2026-06-08 19:04:59,703 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:03,871 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:07,985 - pyscenedetect - INFO - Detecting scenes...
+
 77%|███████▋  | 83/108 [22:19<05:17, 12.69s/it]2026-06-08 19:05:12,145 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:16,705 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:21,223 - pyscenedetect - INFO - Detecting scenes...
+
 78%|███████▊  | 84/108 [22:32<05:07, 12.82s/it]2026-06-08 19:05:25,291 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:29,365 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:34,240 - pyscenedetect - INFO - Detecting scenes...
+
 79%|███████▊  | 85/108 [22:47<05:07, 13.35s/it]2026-06-08 19:05:39,900 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:45,629 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:05:50,583 - pyscenedetect - INFO - Detecting scenes...
+
 80%|███████▉  | 86/108 [23:03<05:12, 14.22s/it]2026-06-08 19:05:56,131 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:06:01,813 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:06:07,105 - pyscenedetect - INFO - Detecting scenes...
+
 81%|████████  | 87/108 [23:21<05:19, 15.20s/it]2026-06-08 19:06:13,614 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:06:20,709 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:06:27,499 - pyscenedetect - INFO - Detecting scenes...
+
 81%|████████▏ | 88/108 [23:41<05:33, 16.67s/it]2026-06-08 19:06:33,729 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:06:41,330 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:06:48,275 - pyscenedetect - INFO - Detecting scenes...
+
 82%|████████▏ | 89/108 [24:02<05:44, 18.14s/it]2026-06-08 19:06:55,308 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:07:01,776 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:07:08,743 - pyscenedetect - INFO - Detecting scenes...
+
 83%|████████▎ | 90/108 [24:23<05:38, 18.81s/it]2026-06-08 19:07:15,648 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:07:22,253 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:07:28,657 - pyscenedetect - INFO - Detecting scenes...
+
 84%|████████▍ | 91/108 [24:43<05:27, 19.24s/it]2026-06-08 19:07:35,909 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:07:43,077 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:07:49,532 - pyscenedetect - INFO - Detecting scenes...
+
 85%|████████▌ | 92/108 [25:04<05:15, 19.70s/it]2026-06-08 19:07:56,697 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:08:03,848 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:08:10,469 - pyscenedetect - INFO - Detecting scenes...
+
 86%|████████▌ | 93/108 [25:25<05:01, 20.07s/it]2026-06-08 19:08:17,630 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:08:24,567 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:08:31,058 - pyscenedetect - INFO - Detecting scenes...
+
 87%|████████▋ | 94/108 [25:45<04:40, 20.06s/it]2026-06-08 19:08:37,676 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:08:44,611 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:08:51,375 - pyscenedetect - INFO - Detecting scenes...
+
 88%|████████▊ | 95/108 [26:05<04:22, 20.18s/it]2026-06-08 19:08:58,120 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:09:05,485 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:09:13,527 - pyscenedetect - INFO - Detecting scenes...
+
 89%|████████▉ | 96/108 [26:27<04:09, 20.79s/it]2026-06-08 19:09:20,348 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:09:27,493 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:09:34,045 - pyscenedetect - INFO - Detecting scenes...
+
 90%|████████▉ | 97/108 [26:48<03:48, 20.78s/it]2026-06-08 19:09:41,097 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:09:48,256 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:09:55,833 - pyscenedetect - INFO - Detecting scenes...
+
 91%|█████████ | 98/108 [27:10<03:31, 21.12s/it]2026-06-08 19:10:02,990 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:10:10,307 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:10:19,166 - pyscenedetect - INFO - Detecting scenes...
+
 92%|█████████▏| 99/108 [27:33<03:15, 21.73s/it]2026-06-08 19:10:26,167 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:10:32,994 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:10:39,921 - pyscenedetect - INFO - Detecting scenes...
+
 93%|█████████▎| 100/108 [27:54<02:51, 21.42s/it]2026-06-08 19:10:46,870 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:10:53,354 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:11:00,137 - pyscenedetect - INFO - Detecting scenes...
+
 94%|█████████▎| 101/108 [28:14<02:28, 21.17s/it]2026-06-08 19:11:07,440 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:11:14,564 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:11:21,064 - pyscenedetect - INFO - Detecting scenes...
+
 94%|█████████▍| 102/108 [28:34<02:05, 20.85s/it]2026-06-08 19:11:27,566 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:11:34,697 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:11:41,797 - pyscenedetect - INFO - Detecting scenes...
+
 95%|█████████▌| 103/108 [28:55<01:44, 20.83s/it]2026-06-08 19:11:48,328 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:11:55,117 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:12:01,770 - pyscenedetect - INFO - Detecting scenes...
+
 96%|█████████▋| 104/108 [29:16<01:23, 20.76s/it]2026-06-08 19:12:08,921 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:12:15,936 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:12:22,930 - pyscenedetect - INFO - Detecting scenes...
+
 97%|█████████▋| 105/108 [29:37<01:02, 20.86s/it]2026-06-08 19:12:30,036 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:12:37,088 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:12:43,817 - pyscenedetect - INFO - Detecting scenes...
+
 98%|█████████▊| 106/108 [29:57<00:41, 20.74s/it]2026-06-08 19:12:50,506 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:12:57,548 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:13:04,699 - pyscenedetect - INFO - Detecting scenes...
+
 99%|█████████▉| 107/108 [30:19<00:20, 20.90s/it]2026-06-08 19:13:11,761 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:13:18,427 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 19:13:23,982 - pyscenedetect - INFO - Detecting scenes...
+
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion/results_2026-06-08-18:42:29_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Complex_Landscape.log

@@ -0,0 +1,24 @@
+[2026-06-08 20:58:52 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Complex_Landscape --dimension Complex_Landscape --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Complex_Landscape', dimension=['Complex_Landscape'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+File exists: /root/ckpts/vbench2/Qwen/Qwen2.5-7B-Instruct
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape/results_2026-06-08-20:58:55_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape/results_2026-06-08-20:58:55_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 20:59:00,237 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+2026-06-08 20:59:05,556 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+
  0%|          | 0/30 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  3%|▎         | 1/30 [00:49<24:02, 49.74s/it]
  7%|▋         | 2/30 [01:33<21:39, 46.40s/it]
 10%|█         | 3/30 [02:20<20:53, 46.42s/it]
 13%|█▎        | 4/30 [03:01<19:10, 44.25s/it]
 17%|█▋        | 5/30 [03:56<20:07, 48.30s/it]
 20%|██        | 6/30 [04:44<19:19, 48.30s/it]
 23%|██▎       | 7/30 [05:33<18:30, 48.27s/it]
 27%|██▋       | 8/30 [06:10<16:22, 44.66s/it]
 30%|███       | 9/30 [06:48<14:59, 42.82s/it]
 33%|███▎      | 10/30 [07:43<15:29, 46.46s/it]
 37%|███▋      | 11/30 [08:42<15:55, 50.31s/it]
 40%|████      | 12/30 [09:28<14:44, 49.12s/it]
 43%|████▎     | 13/30 [10:12<13:24, 47.35s/it]
 47%|████▋     | 14/30 [11:07<13:14, 49.63s/it]
 50%|█████     | 15/30 [11:48<11:48, 47.20s/it]
 53%|█████▎    | 16/30 [12:35<10:58, 47.04s/it]
 57%|█████▋    | 17/30 [13:26<10:28, 48.34s/it]
 60%|██████    | 18/30 [14:14<09:39, 48.28s/it]
 63%|██████▎   | 19/30 [15:01<08:46, 47.85s/it]
 67%|██████▋   | 20/30 [16:01<08:33, 51.35s/it]
 70%|███████   | 21/30 [16:44<07:19, 48.85s/it]
 73%|███████▎  | 22/30 [17:26<06:16, 47.00s/it]
 77%|███████▋  | 23/30 [18:12<05:25, 46.48s/it]
 80%|████████  | 24/30 [18:54<04:31, 45.25s/it]
 83%|████████▎ | 25/30 [19:28<03:29, 41.96s/it]
 87%|████████▋ | 26/30 [20:14<02:51, 42.97s/it]
 90%|█████████ | 27/30 [20:52<02:04, 41.59s/it]
 93%|█████████▎| 28/30 [21:33<01:22, 41.37s/it]
 97%|█████████▋| 29/30 [22:26<00:44, 44.97s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape/results_2026-06-08-20:58:55_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Complex_Plot.log

@@ -0,0 +1,24 @@
+[2026-06-08 19:22:57 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Complex_Plot --dimension Complex_Plot --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Complex_Plot', dimension=['Complex_Plot'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+File exists: /root/ckpts/vbench2/Qwen/Qwen2.5-7B-Instruct
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot/results_2026-06-08-19:23:02_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot/results_2026-06-08-19:23:02_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 19:23:14,518 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+2026-06-08 19:23:22,547 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+
  0%|          | 0/60 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  2%|▏         | 1/60 [00:58<57:17, 58.27s/it]
  3%|▎         | 2/60 [01:47<51:03, 52.82s/it]
  5%|▌         | 3/60 [02:48<53:44, 56.57s/it]
  7%|▋         | 4/60 [03:32<48:23, 51.85s/it]
  8%|▊         | 5/60 [04:01<39:54, 43.53s/it]
 10%|█         | 6/60 [04:33<35:38, 39.61s/it]
 12%|█▏        | 7/60 [05:04<32:22, 36.64s/it]
 13%|█▎        | 8/60 [05:38<31:07, 35.91s/it]
 15%|█▌        | 9/60 [06:14<30:39, 36.07s/it]
 17%|█▋        | 10/60 [06:51<30:03, 36.07s/it]
 18%|█▊        | 11/60 [07:26<29:14, 35.81s/it]
 20%|██        | 12/60 [08:04<29:08, 36.43s/it]
 22%|██▏       | 13/60 [08:44<29:27, 37.61s/it]
 23%|██▎       | 14/60 [09:50<35:18, 46.06s/it]
 25%|██▌       | 15/60 [10:34<34:09, 45.54s/it]
 27%|██▋       | 16/60 [11:39<37:42, 51.41s/it]
 28%|██▊       | 17/60 [12:16<33:49, 47.19s/it]
 30%|███       | 18/60 [13:16<35:45, 51.08s/it]
 32%|███▏      | 19/60 [14:00<33:18, 48.73s/it]
 33%|███▎      | 20/60 [14:46<32:00, 48.00s/it]
 35%|███▌      | 21/60 [15:35<31:22, 48.28s/it]
 37%|███▋      | 22/60 [16:26<31:11, 49.26s/it]
 38%|███▊      | 23/60 [17:01<27:41, 44.90s/it]
 40%|████      | 24/60 [17:37<25:15, 42.11s/it]
 42%|████▏     | 25/60 [18:10<22:58, 39.38s/it]
 43%|████▎     | 26/60 [18:41<20:53, 36.88s/it]
 45%|████▌     | 27/60 [19:22<20:58, 38.14s/it]
 47%|████▋     | 28/60 [20:02<20:35, 38.61s/it]
 48%|████▊     | 29/60 [20:28<18:07, 35.08s/it]
 50%|█████     | 30/60 [20:59<16:49, 33.65s/it]
 52%|█████▏    | 31/60 [21:31<16:01, 33.14s/it]
 53%|█████▎    | 32/60 [22:02<15:14, 32.66s/it]
 55%|█████▌    | 33/60 [22:34<14:32, 32.33s/it]
 57%|█████▋    | 34/60 [23:06<14:01, 32.36s/it]
 58%|█████▊    | 35/60 [23:38<13:23, 32.15s/it]
 60%|██████    | 36/60 [24:07<12:29, 31.22s/it]
 62%|██████▏   | 37/60 [24:39<12:06, 31.61s/it]
 63%|██████▎   | 38/60 [25:19<12:26, 33.93s/it]
 65%|██████▌   | 39/60 [25:54<12:03, 34.45s/it]
 67%|██████▋   | 40/60 [26:41<12:38, 37.93s/it]
 68%|██████▊   | 41/60 [27:19<12:04, 38.16s/it]
 70%|███████   | 42/60 [27:52<10:56, 36.46s/it]
 72%|███████▏  | 43/60 [28:21<09:45, 34.44s/it]
 73%|███████▎  | 44/60 [28:57<09:16, 34.79s/it]
 75%|███████▌  | 45/60 [29:27<08:18, 33.24s/it]
 77%|███████▋  | 46/60 [29:57<07:35, 32.51s/it]
 78%|███████▊  | 47/60 [30:36<07:24, 34.17s/it]
 80%|████████  | 48/60 [31:14<07:03, 35.32s/it]
 82%|████████▏ | 49/60 [31:49<06:30, 35.49s/it]
 83%|████████▎ | 50/60 [32:26<05:57, 35.73s/it]
 85%|████████▌ | 51/60 [32:59<05:13, 34.85s/it]
 87%|████████▋ | 52/60 [33:34<04:40, 35.12s/it]
 88%|████████▊ | 53/60 [34:09<04:04, 34.91s/it]
 90%|█████████ | 54/60 [34:42<03:26, 34.43s/it]
 92%|█████████▏| 55/60 [35:09<02:40, 32.15s/it]
 93%|█████████▎| 56/60 [35:41<02:08, 32.02s/it]
 95%|█████████▌| 57/60 [36:18<01:40, 33.55s/it]
 97%|█████████▋| 58/60 [37:00<01:12, 36.12s/it]
 98%|█████████▊| 59/60 [37:38<00:36, 36.82s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot/results_2026-06-08-19:23:02_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Diversity.log

@@ -0,0 +1,12 @@
+[2026-06-08 17:28:50 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Diversity --dimension Diversity --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Diversity', dimension=['Diversity'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity/results_2026-06-08-17:28:53_full_info.json
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity/results_2026-06-08-17:28:53_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
+  warnings.warn(
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG19_Weights.IMAGENET1K_V1`. You can also use `weights=VGG19_Weights.DEFAULT` to get the most up-to-date weights.
+  warnings.warn(msg)
+
  0%|          | 0/10 [00:00<?, ?it/s]
 10%|█         | 1/10 [13:18<1:59:43, 798.16s/it]
 20%|██        | 2/10 [26:39<1:46:41, 800.21s/it]
 30%|███       | 3/10 [40:11<1:33:58, 805.57s/it]
 40%|████      | 4/10 [53:01<1:19:07, 791.25s/it]
 50%|█████     | 5/10 [1:05:48<1:05:14, 782.83s/it]
 60%|██████    | 6/10 [1:18:49<52:08, 782.22s/it]  
 70%|███████   | 7/10 [1:31:42<38:57, 779.07s/it]
 80%|████████  | 8/10 [1:45:12<26:17, 788.78s/it]
 90%|█████████ | 9/10 [1:58:05<13:03, 783.90s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity/results_2026-06-08-17:28:53_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Dynamic_Attribute.log

@@ -0,0 +1,21 @@
+[2026-06-08 18:53:25 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Dynamic_Attribute --dimension Dynamic_Attribute --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Dynamic_Attribute', dimension=['Dynamic_Attribute'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute/results_2026-06-08-18:53:30_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute/results_2026-06-08-18:53:30_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 18:53:42,588 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+
  0%|          | 0/91 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  1%|          | 1/91 [00:25<38:04, 25.39s/it]
  2%|▏         | 2/91 [00:50<37:07, 25.03s/it]
  3%|▎         | 3/91 [01:14<36:34, 24.94s/it]
  4%|▍         | 4/91 [01:39<35:51, 24.73s/it]
  5%|▌         | 5/91 [02:04<35:27, 24.73s/it]
  7%|▋         | 6/91 [02:28<35:04, 24.76s/it]
  8%|▊         | 7/91 [02:54<34:49, 24.88s/it]
  9%|▉         | 8/91 [03:18<34:13, 24.74s/it]
 10%|▉         | 9/91 [03:42<33:36, 24.59s/it]
 11%|█         | 10/91 [04:07<33:06, 24.53s/it]
 12%|█▏        | 11/91 [04:32<32:52, 24.66s/it]
 13%|█▎        | 12/91 [04:56<32:26, 24.63s/it]
 14%|█▍        | 13/91 [05:21<32:10, 24.75s/it]
 15%|█▌        | 14/91 [05:46<31:47, 24.77s/it]
 16%|█▋        | 15/91 [06:10<31:13, 24.65s/it]
 18%|█▊        | 16/91 [06:35<30:37, 24.50s/it]
 19%|█▊        | 17/91 [06:58<29:40, 24.05s/it]
 20%|█▉        | 18/91 [07:20<28:42, 23.59s/it]
 21%|██        | 19/91 [07:43<27:53, 23.24s/it]
 22%|██▏       | 20/91 [08:06<27:25, 23.17s/it]
 23%|██▎       | 21/91 [08:28<26:54, 23.06s/it]
 24%|██▍       | 22/91 [08:51<26:30, 23.05s/it]
 25%|██▌       | 23/91 [09:14<25:54, 22.86s/it]
 26%|██▋       | 24/91 [09:36<25:25, 22.78s/it]
 27%|���█▋       | 25/91 [09:59<24:56, 22.67s/it]
 29%|██▊       | 26/91 [10:21<24:32, 22.66s/it]
 30%|██▉       | 27/91 [10:44<24:05, 22.58s/it]
 31%|███       | 28/91 [11:06<23:42, 22.58s/it]
 32%|███▏      | 29/91 [11:29<23:19, 22.57s/it]
 33%|███▎      | 30/91 [11:52<23:08, 22.76s/it]
 34%|███▍      | 31/91 [12:16<22:59, 23.00s/it]
 35%|███▌      | 32/91 [12:26<18:55, 19.25s/it]
 36%|███▋      | 33/91 [12:37<16:15, 16.83s/it]
 37%|███▋      | 34/91 [12:49<14:23, 15.15s/it]
 38%|███▊      | 35/91 [13:00<13:03, 13.99s/it]
 40%|███▉      | 36/91 [13:11<12:04, 13.17s/it]
 41%|████      | 37/91 [13:23<11:24, 12.68s/it]
 42%|████▏     | 38/91 [13:34<10:54, 12.35s/it]
 43%|████▎     | 39/91 [13:46<10:24, 12.01s/it]
 44%|████▍     | 40/91 [13:57<10:02, 11.81s/it]
 45%|████▌     | 41/91 [14:08<09:44, 11.68s/it]
 46%|████▌     | 42/91 [14:19<09:19, 11.41s/it]
 47%|████▋     | 43/91 [14:30<09:06, 11.39s/it]
 48%|████▊     | 44/91 [14:42<08:58, 11.46s/it]
 49%|████▉     | 45/91 [14:54<08:51, 11.56s/it]
 51%|█████     | 46/91 [15:05<08:42, 11.61s/it]
 52%|█████▏    | 47/91 [15:17<08:31, 11.62s/it]
 53%|█████▎    | 48/91 [15:29<08:20, 11.63s/it]
 54%|█████▍    | 49/91 [15:40<08:08, 11.64s/it]
 55%|█████▍    | 50/91 [15:52<07:53, 11.56s/it]
 56%|█████▌    | 51/91 [16:03<07:39, 11.48s/it]
 57%|█████▋    | 52/91 [16:15<07:27, 11.49s/it]
 58%|█████▊    | 53/91 [16:26<07:15, 11.45s/it]
 59%|█████▉    | 54/91 [16:38<07:06, 11.52s/it]
 60%|██████    | 55/91 [16:50<06:58, 11.63s/it]
 62%|██████▏   | 56/91 [17:01<06:43, 11.52s/it]
 63%|██████▎   | 57/91 [17:12<06:29, 11.46s/it]
 64%|██████▎   | 58/91 [17:24<06:19, 11.49s/it]
 65%|██████▍   | 59/91 [17:35<06:07, 11.47s/it]
 66%|██████▌   | 60/91 [17:47<06:01, 11.65s/it]
 67%|██████▋   | 61/91 [17:59<05:47, 11.58s/it]
 68%|██████▊   | 62/91 [18:10<05:35, 11.56s/it]
 69%|██████▉   | 63/91 [18:21<05:21, 11.47s/it]
 70%|███████   | 64/91 [18:33<05:09, 11.46s/it]
 71%|███████▏  | 65/91 [18:44<04:57, 11.46s/it]
 73%|███████▎  | 66/91 [18:56<04:46, 11.44s/it]
 74%|███████▎  | 67/91 [19:07<04:33, 11.38s/it]
 75%|███████▍  | 68/91 [19:32<05:55, 15.47s/it]
 76%|███████▌  | 69/91 [19:55<06:32, 17.83s/it]
 77%|███████▋  | 70/91 [20:18<06:46, 19.37s/it]
 78%|███████▊  | 71/91 [20:41<06:46, 20.32s/it]
 79%|███████▉  | 72/91 [21:03<06:38, 20.99s/it]
 80%|████████  | 73/91 [21:26<06:26, 21.45s/it]
 81%|████████▏ | 74/91 [21:48<06:09, 21.74s/it]
 82%|████████▏ | 75/91 [22:11<05:50, 21.91s/it]
 84%|████████▎ | 76/91 [22:33<05:32, 22.14s/it]
 85%|████████▍ | 77/91 [22:56<05:10, 22.18s/it]
 86%|████████▌ | 78/91 [23:18<04:49, 22.24s/it]
 87%|████████▋ | 79/91 [23:40<04:27, 22.29s/it]
 88%|████████▊ | 80/91 [24:03<04:05, 22.33s/it]
 89%|████████▉ | 81/91 [24:25<03:43, 22.33s/it]
 90%|█████████ | 82/91 [24:48<03:22, 22.49s/it]
 91%|█████████ | 83/91 [25:11<03:02, 22.78s/it]
 92%|█████████▏| 84/91 [25:36<02:42, 23.19s/it]
 93%|█████████▎| 85/91 [26:00<02:21, 23.53s/it]
 95%|█████████▍| 86/91 [26:25<01:59, 23.96s/it]
 96%|█████████▌| 87/91 [26:49<01:36, 24.15s/it]
 97%|█████████▋| 88/91 [27:14<01:12, 24.23s/it]
 98%|█████████▊| 89/91 [27:38<00:48, 24.36s/it]
 99%|█████████▉| 90/91 [28:03<00:24, 24.53s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute/results_2026-06-08-18:53:30_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Human_Anatomy.log

@@ -0,0 +1,23 @@
+[2026-06-08 17:28:50 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Anatomy --dimension Human_Anatomy --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Anatomy', dimension=['Human_Anatomy'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy/results_2026-06-08-17:28:53_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy/results_2026-06-08-17:28:53_full_info.json
+Loads checkpoint by local backend from path: /root/ckpts/vbench2/YOLO-World/yolo_world_v2_xl_obj365v1_goldg_cc3mlite_pretrain-5daf1395.pth
+2026-06-08 17:29:02,045 - lvis.lvis - INFO - Loading annotations.
+2026-06-08 17:29:03,315 - lvis.lvis - INFO - Creating index.
+2026-06-08 17:29:03,365 - lvis.lvis - INFO - Index created.
+Loads checkpoint by local backend from path: /root/ckpts/vbench2/YOLO-World/yolo_world_v2_xl_obj365v1_goldg_cc3mlite_pretrain-5daf1395.pth
+2026-06-08 17:29:08,399 - lvis.lvis - INFO - Loading annotations.
+2026-06-08 17:29:09,340 - lvis.lvis - INFO - Creating index.
+2026-06-08 17:29:09,389 - lvis.lvis - INFO - Index created.
+=> merge config from vbench2/third_party/ViTDetector/simmim_finetune__vit_base__img224__800ep.yaml
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/torch/functional.py:505: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /pytorch/aten/src/ATen/native/TensorShape.cpp:4317.)
+  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]
+=> merge config from vbench2/third_party/ViTDetector/simmim_finetune__vit_base__img224__800ep.yaml
+=> merge config from vbench2/third_party/ViTDetector/simmim_finetune__vit_base__img224__800ep.yaml
+
  0%|          | 0/360 [00:00<?, ?it/s]
  0%|          | 1/360 [00:18<1:51:22, 18.61s/it]
  1%|          | 2/360 [00:34<1:41:30, 17.01s/it]
  1%|          | 3/360 [00:47<1:31:25, 15.37s/it]
  1%|          | 4/360 [01:02<1:29:28, 15.08s/it]
  1%|▏         | 5/360 [01:15<1:25:36, 14.47s/it]
  2%|▏         | 6/360 [01:29<1:24:31, 14.33s/it]
  2%|▏         | 7/360 [01:52<1:40:26, 17.07s/it]
  2%|▏         | 8/360 [02:17<1:55:10, 19.63s/it]
  2%|▎         | 9/360 [02:41<2:02:04, 20.87s/it]
  3%|▎         | 10/360 [02:57<1:52:20, 19.26s/it]
  3%|▎         | 11/360 [03:11<1:42:40, 17.65s/it]
  3%|▎         | 12/360 [03:26<1:38:59, 17.07s/it]
  4%|▎         | 13/360 [03:43<1:38:49, 17.09s/it]
  4%|▍         | 14/360 [04:01<1:38:29, 17.08s/it]
  4%|▍         | 15/360 [04:15<1:33:00, 16.18s/it]
  4%|▍         | 16/360 [05:40<3:31:32, 36.90s/it]
  5%|▍         | 17/360 [08:35<7:28:33, 78.47s/it]
  5%|▌         | 18/360 [10:23<8:18:42, 87.49s/it]
  5%|▌         | 19/360 [11:06<7:01:29, 74.16s/it]
  6%|▌         | 20/360 [12:19<6:57:12, 73.63s/it]
  6%|▌         | 21/360 [12:49<5:42:19, 60.59s/it]
  6%|▌         | 22/360 [13:04<4:24:23, 46.93s/it]
  6%|▋         | 23/360 [13:17<3:26:23, 36.75s/it]
  7%|▋         | 24/360 [13:30<2:46:41, 29.77s/it]
  7%|▋         | 25/360 [14:08<2:58:31, 31.98s/it]
  7%|▋         | 26/360 [14:33<2:47:40, 30.12s/it]
  8%|▊         | 27/360 [14:56<2:34:55, 27.92s/it]
  8%|▊         | 28/360 [15:11<2:12:57, 24.03s/it]
  8%|▊         | 29/360 [15:26<1:57:45, 21.35s/it]
  8%|▊         | 30/360 [15:41<1:46:16, 19.32s/it]
  9%|▊         | 31/360 [15:56<1:38:46, 18.01s/it]
  9%|▉         | 32/360 [16:11<1:34:07, 17.22s/it]
  9%|▉         | 33/360 [16:25<1:27:32, 16.06s/it]
  9%|▉         | 34/360 [16:58<1:56:01, 21.35s/it]
 10%|▉         | 35/360 [17:38<2:25:33, 26.87s/it]
 10%|█         | 36/360 [18:24<2:56:17, 32.65s/it]
 10%|█         | 37/360 [19:15<3:25:27, 38.17s/it]
 11%|█         | 38/360 [20:20<4:07:04, 46.04s/it]
 11%|█         | 39/360 [21:10<4:12:53, 47.27s/it]
 11%|█         | 40/360 [22:06<4:27:16, 50.11s/it]
 11%|█▏        | 41/360 [23:09<4:45:42, 53.74s/it]
 12%|█▏        | 42/360 [24:02<4:44:18, 53.64s/it]
 12%|█▏        | 43/360 [25:34<5:44:44, 65.25s/it]
 12%|█▏        | 44/360 [26:24<5:18:35, 60.49s/it]
 12%|█▎        | 45/360 [26:38<4:05:19, 46.73s/it]
 13%|█▎        | 46/360 [27:04<3:30:49, 40.29s/it]
 13%|█▎        | 47/360 [27:27<3:03:46, 35.23s/it]
 13%|█▎        | 48/360 [27:40<2:28:29, 28.56s/it]
 14%|█▎        | 49/360 [27:55<2:06:29, 24.40s/it]
 14%|█▍        | 50/360 [28:14<1:57:48, 22.80s/it]
 14%|█▍        | 51/360 [28:28<1:44:37, 20.32s/it]
 14%|█▍        | 52/360 [28:46<1:40:09, 19.51s/it]
 15%|█▍        | 53/360 [29:04<1:37:37, 19.08s/it]
 15%|█▌        | 54/360 [29:35<1:54:53, 22.53s/it]
 15%|█▌        | 55/360 [29:49<1:42:47, 20.22s/it]
 16%|█▌        | 56/360 [30:15<1:50:37, 21.83s/it]
 16%|█▌        | 57/360 [30:35<1:47:57, 21.38s/it]
 16%|█▌        | 58/360 [30:49<1:35:59, 19.07s/it]
 16%|█▋        | 59/360 [31:04<1:28:45, 17.69s/it]
 17%|█▋        | 60/360 [31:18<1:23:47, 16.76s/it]
 17%|█▋        | 61/360 [31:30<1:16:17, 15.31s/it]
 17%|█▋        | 62/360 [31:44<1:13:43, 14.84s/it]
 18%|█▊        | 63/360 [32:37<2:10:09, 26.30s/it]
 18%|█▊        | 64/360 [33:44<3:10:53, 38.69s/it]
 18%|█▊        | 65/360 [34:47<3:45:04, 45.78s/it]
 18%|█▊        | 66/360 [35:41<3:57:28, 48.47s/it]
 19%|█▊        | 67/360 [36:26<3:50:37, 47.23s/it]
 19%|█▉        | 68/360 [36:59<3:28:47, 42.90s/it]
 19%|█▉        | 69/360 [38:23<4:28:50, 55.43s/it]
 19%|█▉        | 70/360 [39:23<4:34:25, 56.78s/it]
 20%|█▉        | 71/360 [39:52<3:53:07, 48.40s/it]
 20%|██        | 72/360 [40:08<3:04:59, 38.54s/it]
 20%|██        | 73/360 [40:20<2:27:31, 30.84s/it]
 21%|██        | 74/360 [40:44<2:16:36, 28.66s/it]
 21%|██        | 75/360 [41:00<1:58:04, 24.86s/it]
 21%|██        | 76/360 [41:21<1:51:29, 23.55s/it]
 21%|██▏       | 77/360 [41:41<1:46:40, 22.62s/it]
 22%|██▏       | 78/360 [41:56<1:36:16, 20.49s/it]
 22%|██▏       | 79/360 [42:11<1:26:52, 18.55s/it]
 22%|██▏       | 80/360 [42:23<1:18:10, 16.75s/it]
 22%|██▎       | 81/360 [42:35<1:10:59, 15.27s/it]
 23%|██▎       | 82/360 [42:47<1:06:08, 14.28s/it]
 23%|██▎       | 83/360 [43:00<1:03:59, 13.86s/it]
 23%|██▎       | 84/360 [43:17<1:08:23, 14.87s/it]
 24%|██▎       | 85/360 [43:33<1:09:05, 15.07s/it]
 24%|██▍       | 86/360 [44:14<1:45:00, 22.99s/it]
 24%|██▍       | 87/360 [44:29<1:33:35, 20.57s/it]
 24%|██▍       | 88/360 [44:42<1:23:14, 18.36s/it]
 25%|██▍       | 89/360 [45:29<2:01:07, 26.82s/it]
 25%|██▌       | 90/360 [46:22<2:36:49, 34.85s/it]
 25%|██▌       | 91/360 [47:01<2:41:48, 36.09s/it]
 26%|██▌       | 92/360 [47:51<2:59:52, 40.27s/it]
 26%|██▌       | 93/360 [48:43<3:14:57, 43.81s/it]
 26%|██▌       | 94/360 [50:26<4:32:05, 61.37s/it]
 26%|██▋       | 95/360 [51:55<5:07:58, 69.73s/it]
 27%|██▋       | 96/360 [52:41<4:35:13, 62.55s/it]
 27%|██▋       | 97/360 [52:54<3:28:49, 47.64s/it]
 27%|██▋       | 98/360 [53:04<2:39:40, 36.57s/it]
 28%|██▊       | 99/360 [53:15<2:05:03, 28.75s/it]
 28%|██▊       | 100/360 [53:29<1:45:05, 24.25s/it]
 28%|██▊       | 101/360 [53:42<1:30:54, 21.06s/it]
 28%|██▊       | 102/360 [53:56<1:20:40, 18.76s/it]
 29%|██▊       | 103/360 [54:10<1:14:10, 17.32s/it]
 29%|██▉       | 104/360 [54:22<1:07:40, 15.86s/it]
 29%|██▉       | 105/360 [54:34<1:02:18, 14.66s/it]
 29%|██▉       | 106/360 [54:47<59:59, 14.17s/it]  
 30%|██▉       | 107/360 [54:59<57:18, 13.59s/it]
 30%|███       | 108/360 [55:12<56:00, 13.33s/it]
 30%|███       | 109/360 [55:24<54:26, 13.01s/it]
 31%|███       | 110/360 [55:35<51:47, 12.43s/it]
 31%|███       | 111/360 [55:48<51:58, 12.53s/it]
 31%|███       | 112/360 [55:55<45:11, 10.93s/it]
 31%|███▏      | 113/360 [56:01<39:09,  9.51s/it]
 32%|███▏      | 114/360 [56:13<41:11, 10.05s/it]
 32%|███▏      | 115/360 [56:25<44:24, 10.87s/it]
 32%|███▏      | 116/360 [56:37<45:30, 11.19s/it]
 32%|███▎      | 117/360 [56:57<55:19, 13.66s/it]
 33%|███▎      | 118/360 [57:19<1:04:58, 16.11s/it]
 33%|███▎      | 119/360 [58:23<2:02:19, 30.45s/it]
 33%|███▎      | 120/360 [59:20<2:33:53, 38.47s/it]
 34%|███▎      | 121/360 [1:00:05<2:41:40, 40.59s/it]
 34%|███▍      | 122/360 [1:00:34<2:26:56, 37.04s/it]
 34%|███▍      | 123/360 [1:01:18<2:34:31, 39.12s/it]
 34%|███▍      | 124/360 [1:01:53<2:28:29, 37.75s/it]
 35%|███▍      | 125/360 [1:02:16<2:11:27, 33.56s/it]
 35%|███▌      | 126/360 [1:02:36<1:54:04, 29.25s/it]
 35%|███▌      | 127/360 [1:03:23<2:14:37, 34.67s/it]
 36%|███▌      | 128/360 [1:04:08<2:26:24, 37.86s/it]
 36%|███▌      | 129/360 [1:04:53<2:33:25, 39.85s/it]
 36%|███▌      | 130/360 [1:05:29<2:28:42, 38.79s/it]
 36%|███▋      | 131/360 [1:05:46<2:03:00, 32.23s/it]
 37%|███▋      | 132/360 [1:06:04<1:46:44, 28.09s/it]
 37%|███▋      | 133/360 [1:06:17<1:28:21, 23.36s/it]
 37%|███▋      | 134/360 [1:06:29<1:15:28, 20.04s/it]
 38%|███▊      | 135/360 [1:06:42<1:07:42, 18.05s/it]
 38%|███▊      | 136/360 [1:06:58<1:05:01, 17.42s/it]
 38%|███▊      | 137/360 [1:07:07<55:23, 14.90s/it]  
 38%|███▊      | 138/360 [1:07:24<57:35, 15.57s/it]
 39%|███▊      | 139/360 [1:07:39<56:09, 15.25s/it]
 39%|███▉      | 140/360 [1:07:54<55:45, 15.20s/it]
 39%|███▉      | 141/360 [1:08:09<55:21, 15.17s/it]
 39%|███▉      | 142/360 [1:08:25<55:58, 15.41s/it]
 40%|███▉      | 143/360 [1:08:40<55:15, 15.28s/it]
 40%|████      | 144/360 [1:08:51<50:17, 13.97s/it]
 40%|████      | 145/360 [1:09:03<47:31, 13.26s/it]
 41%|████      | 146/360 [1:09:13<44:10, 12.38s/it]
 41%|████      | 147/360 [1:09:30<49:12, 13.86s/it]
 41%|████      | 148/360 [1:09:43<47:48, 13.53s/it]
 41%|████▏     | 149/360 [1:09:55<46:23, 13.19s/it]
 42%|████▏     | 150/360 [1:10:14<51:50, 14.81s/it]
 42%|████▏     | 151/360 [1:10:52<1:16:18, 21.91s/it]
 42%|████▏     | 152/360 [1:11:24<1:25:47, 24.75s/it]
 42%|████▎     | 153/360 [1:12:07<1:44:25, 30.27s/it]
 43%|████▎     | 154/360 [1:12:38<1:45:00, 30.58s/it]
 43%|████▎     | 155/360 [1:13:32<2:08:08, 37.50s/it]
 43%|████▎     | 156/360 [1:14:19<2:17:02, 40.30s/it]
 44%|████▎     | 157/360 [1:15:11<2:28:21, 43.85s/it]
 44%|████▍     | 158/360 [1:15:56<2:28:28, 44.10s/it]
 44%|████▍     | 159/360 [1:16:37<2:25:04, 43.31s/it]
 44%|████▍     | 160/360 [1:17:26<2:30:22, 45.11s/it]
 45%|████▍     | 161/360 [1:18:08<2:26:23, 44.14s/it]
 45%|████▌     | 162/360 [1:18:30<2:03:01, 37.28s/it]
 45%|████▌     | 163/360 [1:18:42<1:38:16, 29.93s/it]
 46%|████▌     | 164/360 [1:18:57<1:22:37, 25.29s/it]
 46%|████▌     | 165/360 [1:19:10<1:10:32, 21.71s/it]
 46%|████▌     | 166/360 [1:19:23<1:01:37, 19.06s/it]
 46%|████▋     | 167/360 [1:19:35<54:31, 16.95s/it]  
 47%|████▋     | 168/360 [1:19:48<50:07, 15.66s/it]
 47%|████▋     | 169/360 [1:19:58<45:09, 14.19s/it]
 47%|████▋     | 170/360 [1:20:10<42:53, 13.55s/it]
 48%|████▊     | 171/360 [1:20:22<40:50, 12.96s/it]
 48%|████▊     | 172/360 [1:20:35<40:34, 12.95s/it]
 48%|████▊     | 173/360 [1:20:50<42:42, 13.71s/it]
 48%|████▊     | 174/360 [1:21:07<44:58, 14.51s/it]
 49%|████▊     | 175/360 [1:21:22<44:54, 14.57s/it]
 49%|████▉     | 176/360 [1:21:38<46:21, 15.12s/it]
 49%|████▉     | 177/360 [1:21:48<41:12, 13.51s/it]
 49%|████▉     | 178/360 [1:22:09<48:09, 15.88s/it]
 50%|████▉     | 179/360 [1:22:26<49:14, 16.32s/it]
 50%|█████     | 180/360 [1:22:44<49:49, 16.61s/it]
 50%|█████     | 181/360 [1:23:00<49:16, 16.52s/it]
 51%|█████     | 182/360 [1:23:21<53:01, 17.88s/it]
 51%|█████     | 183/360 [1:23:54<1:06:02, 22.39s/it]
 51%|█████     | 184/360 [1:24:41<1:27:24, 29.80s/it]
 51%|█████▏    | 185/360 [1:25:25<1:39:06, 33.98s/it]
 52%|█████▏    | 186/360 [1:26:00<1:39:09, 34.19s/it]
 52%|█████▏    | 187/360 [1:26:40<1:43:37, 35.94s/it]
 52%|█████▏    | 188/360 [1:27:23<1:49:11, 38.09s/it]
 52%|█████▎    | 189/360 [1:28:03<1:50:14, 38.68s/it]
 53%|█████▎    | 190/360 [1:28:46<1:53:10, 39.94s/it]
 53%|█████▎    | 191/360 [1:29:25<1:51:57, 39.75s/it]
 53%|█████▎    | 192/360 [1:30:06<1:52:23, 40.14s/it]
 54%|█████▎    | 193/360 [1:30:42<1:48:37, 39.03s/it]
 54%|█████▍    | 194/360 [1:31:09<1:38:03, 35.44s/it]
 54%|█████▍    | 195/360 [1:31:47<1:38:55, 35.97s/it]
 54%|█████▍    | 196/360 [1:32:04<1:23:22, 30.50s/it]
 55%|█████▍    | 197/360 [1:32:26<1:15:18, 27.72s/it]
 55%|█████▌    | 198/360 [1:32:46<1:08:38, 25.42s/it]
 55%|█████▌    | 199/360 [1:33:03<1:01:31, 22.93s/it]
 56%|█████▌    | 200/360 [1:33:17<54:22, 20.39s/it]  
 56%|█████▌    | 201/360 [1:33:33<50:33, 19.08s/it]
 56%|█████▌    | 202/360 [1:33:46<45:09, 17.15s/it]
 56%|█████▋    | 203/360 [1:33:59<41:59, 16.05s/it]
 57%|█████▋    | 204/360 [1:34:13<39:26, 15.17s/it]
 57%|█████▋    | 205/360 [1:34:27<38:30, 14.91s/it]
 57%|█████▋    | 206/360 [1:34:40<36:40, 14.29s/it]
 57%|█████▊    | 207/360 [1:34:54<36:48, 14.44s/it]
 58%|█████▊    | 208/360 [1:34:59<28:56, 11.43s/it]
 58%|█████▊    | 209/360 [1:35:14<31:36, 12.56s/it]
 58%|█████▊    | 210/360 [1:35:32<35:46, 14.31s/it]
 59%|█████▊    | 211/360 [1:35:47<35:23, 14.25s/it]
 59%|█████▉    | 212/360 [1:36:01<35:33, 14.42s/it]
 59%|█████▉    | 213/360 [1:36:26<42:56, 17.53s/it]
 59%|█████▉    | 214/360 [1:36:48<45:57, 18.88s/it]
 60%|█████▉    | 215/360 [1:37:28<1:01:06, 25.29s/it]
 60%|██████    | 216/360 [1:37:48<56:24, 23.50s/it]  
 60%|██████    | 217/360 [1:38:33<1:11:35, 30.04s/it]
 61%|██████    | 218/360 [1:39:25<1:26:32, 36.57s/it]
 61%|██████    | 219/360 [1:40:42<1:54:32, 48.74s/it]
 61%|██████    | 220/360 [1:41:39<1:59:21, 51.15s/it]
 61%|██████▏   | 221/360 [1:42:58<2:18:15, 59.68s/it]
 62%|██████▏   | 222/360 [1:43:50<2:11:29, 57.17s/it]
 62%|██████▏   | 223/360 [1:44:14<1:47:43, 47.18s/it]
 62%|██████▏   | 224/360 [1:44:31<1:26:32, 38.18s/it]
 62%|██████▎   | 225/360 [1:44:48<1:12:00, 32.01s/it]
 63%|██████▎   | 226/360 [1:45:02<59:00, 26.42s/it]  
 63%|██████▎   | 227/360 [1:45:16<50:36, 22.83s/it]
 63%|██████▎   | 228/360 [1:45:30<44:22, 20.17s/it]
 64%|██████▎   | 229/360 [1:45:43<39:27, 18.07s/it]
 64%|██████▍   | 230/360 [1:45:57<35:58, 16.60s/it]
 64%|██████▍   | 231/360 [1:46:12<35:13, 16.39s/it]
 64%|██████▍   | 232/360 [1:46:26<33:16, 15.60s/it]
 65%|██████▍   | 233/360 [1:46:40<31:45, 15.00s/it]
 65%|██████▌   | 234/360 [1:46:56<32:14, 15.35s/it]
 65%|██████▌   | 235/360 [1:47:11<31:46, 15.25s/it]
 66%|██████▌   | 236/360 [1:47:31<34:19, 16.61s/it]
 66%|██████▌   | 237/360 [1:47:46<33:05, 16.15s/it]
 66%|██████▌   | 238/360 [1:47:58<30:27, 14.98s/it]
 66%|██████▋   | 239/360 [1:48:10<28:06, 13.94s/it]
 67%|██████▋   | 240/360 [1:48:22<27:01, 13.51s/it]
 67%|██████▋   | 241/360 [1:48:35<26:23, 13.31s/it]
 67%|██████▋   | 242/360 [1:48:47<25:15, 12.84s/it]
 68%|██████▊   | 243/360 [1:48:57<23:49, 12.22s/it]
 68%|██████▊   | 244/360 [1:49:16<27:25, 14.18s/it]
 68%|██████▊   | 245/360 [1:49:34<29:33, 15.42s/it]
 68%|██████▊   | 246/360 [1:49:56<32:40, 17.20s/it]
 69%|██████▊   | 247/360 [1:50:54<55:27, 29.44s/it]
 69%|██████▉   | 248/360 [1:51:37<1:02:30, 33.49s/it]
 69%|██████▉   | 249/360 [1:52:30<1:13:10, 39.56s/it]
 69%|██████▉   | 250/360 [1:53:14<1:14:30, 40.64s/it]
 70%|██████▉   | 251/360 [1:53:39<1:05:36, 36.12s/it]
 70%|███████   | 252/360 [1:54:23<1:09:07, 38.41s/it]
 70%|███████   | 253/360 [1:54:58<1:06:49, 37.47s/it]
 71%|███████   | 254/360 [1:55:36<1:06:33, 37.67s/it]
 71%|███████   | 255/360 [1:56:15<1:06:27, 37.97s/it]
 71%|███████   | 256/360 [1:56:59<1:08:40, 39.62s/it]
 71%|███████▏  | 257/360 [1:57:32<1:04:51, 37.78s/it]
 72%|███████▏  | 258/360 [1:57:56<57:25, 33.78s/it]  
 72%|███████▏  | 259/360 [1:58:12<47:33, 28.26s/it]
 72%|███████▏  | 260/360 [1:58:22<37:50, 22.70s/it]
 72%|███████▎  | 261/360 [1:58:41<36:01, 21.83s/it]
 73%|███████▎  | 262/360 [1:59:05<36:28, 22.33s/it]
 73%|███████▎  | 263/360 [1:59:25<35:04, 21.70s/it]
 73%|███████▎  | 264/360 [1:59:45<33:58, 21.23s/it]
 74%|███████▎  | 265/360 [2:00:14<36:57, 23.34s/it]
 74%|███████▍  | 266/360 [2:00:36<36:23, 23.23s/it]
 74%|███████▍  | 267/360 [2:00:50<31:29, 20.32s/it]
 74%|███████▍  | 268/360 [2:01:07<29:31, 19.25s/it]
 75%|███████▍  | 269/360 [2:01:24<28:11, 18.59s/it]
 75%|███████▌  | 270/360 [2:01:45<28:51, 19.24s/it]
 75%|███████▌  | 271/360 [2:02:03<28:08, 18.97s/it]
 76%|███████▌  | 272/360 [2:02:58<43:55, 29.95s/it]
 76%|███████▌  | 273/360 [2:04:06<59:44, 41.20s/it]
 76%|███████▌  | 274/360 [2:04:59<1:04:05, 44.71s/it]
 76%|███████▋  | 275/360 [2:05:26<56:01, 39.54s/it]  
 77%|███████▋  | 276/360 [2:06:44<1:11:19, 50.94s/it]
 77%|███████▋  | 277/360 [2:07:30<1:08:29, 49.51s/it]
 77%|███████▋  | 278/360 [2:08:11<1:04:03, 46.87s/it]
 78%|███████▊  | 279/360 [2:08:57<1:03:14, 46.84s/it]
 78%|███████▊  | 280/360 [2:09:55<1:06:39, 49.99s/it]
 78%|██��████▊  | 281/360 [2:10:36<1:02:29, 47.46s/it]
 78%|███████▊  | 282/360 [2:11:02<53:19, 41.02s/it]  
 79%|███████▊  | 283/360 [2:11:19<43:13, 33.69s/it]
 79%|███████▉  | 284/360 [2:11:34<35:38, 28.14s/it]
 79%|███████▉  | 285/360 [2:11:53<31:45, 25.41s/it]
 79%|███████▉  | 286/360 [2:12:13<29:26, 23.87s/it]
 80%|███████▉  | 287/360 [2:12:33<27:22, 22.50s/it]
 80%|████████  | 288/360 [2:12:50<25:15, 21.05s/it]
 80%|████████  | 289/360 [2:13:03<21:43, 18.36s/it]
 81%|████████  | 290/360 [2:13:13<18:48, 16.13s/it]
 81%|████████  | 291/360 [2:13:25<16:56, 14.73s/it]
 81%|████████  | 292/360 [2:13:36<15:19, 13.53s/it]
 81%|████████▏ | 293/360 [2:13:48<14:46, 13.22s/it]
 82%|████████▏ | 294/360 [2:14:01<14:20, 13.04s/it]
 82%|████████▏ | 295/360 [2:14:19<15:39, 14.45s/it]
 82%|████████▏ | 296/360 [2:14:35<16:10, 15.17s/it]
 82%|████████▎ | 297/360 [2:14:52<16:31, 15.73s/it]
 83%|████████▎ | 298/360 [2:15:09<16:30, 15.97s/it]
 83%|████████▎ | 299/360 [2:15:23<15:40, 15.42s/it]
 83%|████████▎ | 300/360 [2:15:34<14:04, 14.07s/it]
 84%|████████▎ | 301/360 [2:15:51<14:48, 15.07s/it]
 84%|████████▍ | 302/360 [2:16:11<15:53, 16.45s/it]
 84%|████████▍ | 303/360 [2:16:32<16:52, 17.76s/it]
 84%|████████▍ | 304/360 [2:16:43<14:40, 15.73s/it]
 85%|████████▍ | 305/360 [2:16:54<13:02, 14.22s/it]
 85%|████████▌ | 306/360 [2:17:06<12:19, 13.70s/it]
 85%|████████▌ | 307/360 [2:17:20<12:06, 13.71s/it]
 86%|████████▌ | 308/360 [2:17:35<12:11, 14.07s/it]
 86%|████████▌ | 309/360 [2:17:49<11:58, 14.08s/it]
 86%|████████▌ | 310/360 [2:18:06<12:29, 14.99s/it]
 86%|████████▋ | 311/360 [2:18:23<12:51, 15.75s/it]
 87%|████████▋ | 312/360 [2:18:41<12:55, 16.16s/it]
 87%|████████▋ | 313/360 [2:18:54<12:07, 15.48s/it]
 87%|████████▋ | 314/360 [2:19:15<12:59, 16.94s/it]
 88%|████████▊ | 315/360 [2:19:24<10:59, 14.65s/it]
 88%|████████▊ | 316/360 [2:19:42<11:30, 15.70s/it]
 88%|████████▊ | 317/360 [2:20:00<11:46, 16.43s/it]
 88%|████████▊ | 318/360 [2:20:21<12:23, 17.69s/it]
 89%|████████▊ | 319/360 [2:20:37<11:43, 17.16s/it]
 89%|████████▉ | 320/360 [2:20:45<09:32, 14.30s/it]
 89%|████████▉ | 321/360 [2:20:59<09:14, 14.21s/it]
 89%|████████▉ | 322/360 [2:21:16<09:37, 15.19s/it]
 90%|████████▉ | 323/360 [2:21:34<09:49, 15.94s/it]
 90%|█████████ | 324/360 [2:21:52<09:59, 16.66s/it]
 90%|█████████ | 325/360 [2:22:06<09:10, 15.72s/it]
 91%|█████████ | 326/360 [2:22:20<08:36, 15.18s/it]
 91%|█████████ | 327/360 [2:22:33<08:07, 14.76s/it]
 91%|█████████ | 328/360 [2:22:51<08:18, 15.58s/it]
 91%|█████████▏| 329/360 [2:23:10<08:40, 16.78s/it]
 92%|█████████▏| 330/360 [2:23:32<09:10, 18.36s/it]
 92%|█████████▏| 331/360 [2:23:48<08:26, 17.46s/it]
 92%|█████████▏| 332/360 [2:24:06<08:11, 17.56s/it]
 92%|█████████▎| 333/360 [2:24:25<08:05, 17.98s/it]
 93%|█████████▎| 334/360 [2:24:43<07:49, 18.04s/it]
 93%|█████████▎| 335/360 [2:25:00<07:25, 17.83s/it]
 93%|█████████▎| 336/360 [2:25:20<07:22, 18.45s/it]
 94%|█████████▎| 337/360 [2:25:36<06:46, 17.68s/it]
 94%|█████████▍| 338/360 [2:25:51<06:09, 16.79s/it]
 94%|█████████▍| 339/360 [2:26:04<05:32, 15.86s/it]
 94%|█████████▍| 340/360 [2:26:23<05:37, 16.85s/it]
 95%|█████████▍| 341/360 [2:26:40<05:20, 16.88s/it]
 95%|█████████▌| 342/360 [2:27:00<05:16, 17.57s/it]
 95%|█████████▌| 343/360 [2:27:16<04:50, 17.12s/it]
 96%|█████████▌| 344/360 [2:27:29<04:16, 16.03s/it]
 96%|█████████▌| 345/360 [2:27:47<04:06, 16.46s/it]
 96%|█████████▌| 346/360 [2:28:02<03:44, 16.03s/it]
 96%|█████████▋| 347/360 [2:28:16<03:22, 15.55s/it]
 97%|█████████▋| 348/360 [2:28:29<02:56, 14.71s/it]
 97%|█████████▋| 349/360 [2:28:46<02:50, 15.54s/it]
 97%|█████████▋| 350/360 [2:29:07<02:49, 16.96s/it]
 98%|█████████▊| 351/360 [2:29:22<02:28, 16.48s/it]
 98%|█████████▊| 352/360 [2:29:41<02:19, 17.40s/it]
 98%|█████████▊| 353/360 [2:30:00<02:03, 17.64s/it]
 98%|█████████▊| 354/360 [2:30:18<01:46, 17.75s/it]
 99%|█████████▊| 355/360 [2:30:36<01:29, 17.91s/it]
 99%|█████████▉| 356/360 [2:30:59<01:17, 19.36s/it]
 99%|█████████▉| 357/360 [2:31:13<00:53, 17.81s/it]
 99%|█████████▉| 358/360 [2:31:32<00:36, 18.11s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy/results_2026-06-08-17:28:53_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Human_Clothes.log

@@ -0,0 +1,23 @@
+[2026-06-08 20:23:50 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Clothes --dimension Human_Clothes --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Clothes', dimension=['Human_Clothes'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes/results_2026-06-08-20:23:53_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes/results_2026-06-08-20:23:53_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 20:23:57,997 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+
  0%|          | 0/75 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  1%|▏         | 1/75 [00:22<28:13, 22.88s/it]
  3%|▎         | 2/75 [00:40<24:18, 19.97s/it]
  4%|▍         | 3/75 [01:03<25:20, 21.12s/it]
  5%|▌         | 4/75 [01:25<25:40, 21.69s/it]
  7%|▋         | 5/75 [01:48<25:39, 22.00s/it]
  8%|▊         | 6/75 [02:10<25:30, 22.18s/it]
  9%|▉         | 7/75 [02:33<25:16, 22.30s/it]
 11%|█         | 8/75 [02:56<25:00, 22.39s/it]
 12%|█▏        | 9/75 [03:18<24:41, 22.45s/it]
 13%|█▎        | 10/75 [03:41<24:22, 22.50s/it]
 15%|█▍        | 11/75 [04:03<24:02, 22.54s/it]
 16%|█▌        | 12/75 [04:26<23:42, 22.58s/it]
 17%|█▋        | 13/75 [04:49<23:18, 22.56s/it]
 19%|█▊        | 14/75 [05:11<22:55, 22.54s/it]
 20%|██        | 15/75 [05:34<22:32, 22.55s/it]
 21%|██▏       | 16/75 [05:56<22:11, 22.58s/it]
 23%|██▎       | 17/75 [06:06<17:58, 18.60s/it]
 24%|██▍       | 18/75 [06:24<17:32, 18.46s/it]
 25%|██▌       | 19/75 [06:42<17:06, 18.34s/it]
 27%|██▋       | 20/75 [07:04<17:56, 19.58s/it]
 28%|██▊       | 21/75 [07:27<18:24, 20.46s/it]
 29%|██▉       | 22/75 [07:49<18:36, 21.07s/it]
 31%|███       | 23/75 [08:12<18:41, 21.56s/it]
 32%|███▏      | 24/75 [08:35<18:35, 21.88s/it]
 33%|███▎      | 25/75 [08:57<18:23, 22.07s/it]
 35%|███▍      | 26/75 [09:20<18:08, 22.21s/it]
 36%|███▌      | 27/75 [09:42<17:51, 22.32s/it]
 37%|███▋      | 28/75 [10:05<17:33, 22.41s/it]
 39%|███▊      | 29/75 [10:27<17:12, 22.45s/it]
 40%|████      | 30/75 [10:50<16:51, 22.48s/it]
 41%|████▏     | 31/75 [11:12<16:28, 22.47s/it]
 43%|████▎     | 32/75 [11:31<15:11, 21.19s/it]
 44%|████▍     | 33/75 [11:53<15:07, 21.60s/it]
 45%|████▌     | 34/75 [12:11<14:03, 20.57s/it]
 47%|████▋     | 35/75 [12:34<14:05, 21.15s/it]
 48%|████▊     | 36/75 [12:56<14:01, 21.59s/it]
 49%|████▉     | 37/75 [13:19<13:50, 21.85s/it]
 51%|█████     | 38/75 [13:42<13:36, 22.07s/it]
 52%|█████▏    | 39/75 [14:04<13:19, 22.20s/it]
 53%|█████▎    | 40/75 [14:27<13:00, 22.30s/it]
 55%|█████▍    | 41/75 [14:49<12:40, 22.37s/it]
 56%|█████▌    | 42/75 [15:12<12:19, 22.42s/it]
 57%|█████▋    | 43/75 [15:34<11:58, 22.45s/it]
 59%|█████▊    | 44/75 [15:57<11:35, 22.44s/it]
 60%|██████    | 45/75 [16:19<11:13, 22.47s/it]
 61%|██████▏   | 46/75 [16:37<10:13, 21.17s/it]
 63%|██████▎   | 47/75 [16:55<09:27, 20.27s/it]
 64%|██████▍   | 48/75 [17:18<09:25, 20.95s/it]
 65%|██████▌   | 49/75 [17:40<09:16, 21.41s/it]
 67%|██████▋   | 50/75 [18:03<09:04, 21.77s/it]
 68%|██████▊   | 51/75 [18:26<08:47, 21.98s/it]
 69%|██████▉   | 52/75 [18:48<08:28, 22.13s/it]
 71%|███████   | 53/75 [19:06<07:40, 20.95s/it]
 72%|███████▏  | 54/75 [19:29<07:30, 21.44s/it]
 73%|███████▎  | 55/75 [19:51<07:15, 21.79s/it]
 75%|███████▍  | 56/75 [20:14<06:58, 22.01s/it]
 76%|███████▌  | 57/75 [20:36<06:39, 22.18s/it]
 77%|███████▋  | 58/75 [20:59<06:18, 22.29s/it]
 79%|███████▊  | 59/75 [21:21<05:57, 22.35s/it]
 80%|████████  | 60/75 [21:44<05:36, 22.43s/it]
 81%|████████▏ | 61/75 [22:07<05:14, 22.47s/it]
 83%|████████▎ | 62/75 [22:29<04:52, 22.50s/it]
 84%|████████▍ | 63/75 [22:52<04:29, 22.50s/it]
 85%|████████▌ | 64/75 [23:14<04:07, 22.50s/it]
 87%|████████▋ | 65/75 [23:32<03:31, 21.18s/it]
 88%|████████▊ | 66/75 [23:55<03:14, 21.58s/it]
 89%|████████▉ | 67/75 [24:17<02:55, 21.90s/it]
 91%|█████████ | 68/75 [24:40<02:34, 22.09s/it]
 92%|█████████▏| 69/75 [25:03<02:13, 22.23s/it]
 93%|█████████▎| 70/75 [25:25<01:51, 22.33s/it]
 95%|█████████▍| 71/75 [25:48<01:29, 22.39s/it]
 96%|█████████▌| 72/75 [26:06<01:03, 21.14s/it]
 97%|█████████▋| 73/75 [26:29<00:43, 21.59s/it]
 99%|█████████▊| 74/75 [26:47<00:20, 20.56s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes/results_2026-06-08-20:23:53_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Human_Identity.log

@@ -0,0 +1,14 @@
+[2026-06-08 20:25:20 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Identity --dimension Human_Identity --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Human_Identity', dimension=['Human_Identity'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity/results_2026-06-08-20:25:23_full_info.json
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity/results_2026-06-08-20:25:23_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/torch/hub.py:879: FutureWarning: Falling back to the old format < 1.6. This support will be deprecated in favor of default zipfile format introduced in 1.6. Please redo torch.save() to save it in the new zipfile format.
+  return _legacy_zip_load(cached_file, model_dir, map_location, weights_only)
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
+  warnings.warn(
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.
+  warnings.warn(msg)
+
  0%|          | 0/46 [00:00<?, ?it/s]
  2%|▏         | 1/46 [01:11<53:39, 71.55s/it]
  4%|▍         | 2/46 [01:59<42:13, 57.59s/it]
  7%|▋         | 3/46 [03:09<45:13, 63.10s/it]
  9%|▊         | 4/46 [04:18<46:00, 65.74s/it]
 11%|█         | 5/46 [05:28<45:47, 67.01s/it]
 13%|█▎        | 6/46 [06:38<45:21, 68.05s/it]
 15%|█▌        | 7/46 [07:03<35:11, 54.15s/it]
 17%|█▋        | 8/46 [07:52<33:10, 52.39s/it]
 20%|█▉        | 9/46 [08:18<27:15, 44.20s/it]
 22%|██▏       | 10/46 [09:07<27:28, 45.79s/it]
 24%|██▍       | 11/46 [09:57<27:22, 46.94s/it]
 26%|██▌       | 12/46 [10:46<27:02, 47.72s/it]
 28%|██▊       | 13/46 [11:59<30:24, 55.30s/it]
 30%|███       | 14/46 [12:48<28:29, 53.41s/it]
 33%|███▎      | 15/46 [12:51<19:47, 38.29s/it]
 35%|███▍      | 16/46 [12:55<13:52, 27.74s/it]
 37%|███▋      | 17/46 [13:21<13:10, 27.26s/it]
 39%|███▉      | 18/46 [14:34<19:06, 40.96s/it]
 41%|████▏     | 19/46 [15:46<22:38, 50.31s/it]
 43%|████▎     | 20/46 [15:49<15:39, 36.13s/it]
 46%|████▌     | 21/46 [16:15<13:48, 33.12s/it]
 48%|████▊     | 22/46 [16:41<12:27, 31.15s/it]
 50%|█████     | 23/46 [16:44<08:42, 22.70s/it]
 52%|█████▏    | 24/46 [17:34<11:14, 30.66s/it]
 54%|█████▍    | 25/46 [18:46<15:07, 43.22s/it]
 57%|█████▋    | 26/46 [19:35<14:58, 44.93s/it]
 59%|█████▊    | 27/46 [20:25<14:42, 46.42s/it]
 61%|██████    | 28/46 [20:51<12:07, 40.42s/it]
 63%|██████▎   | 29/46 [20:55<08:16, 29.23s/it]
 65%|██████▌   | 30/46 [21:44<09:23, 35.24s/it]
 67%|██████▋   | 31/46 [22:33<09:50, 39.39s/it]
 70%|██████▉   | 32/46 [22:59<08:15, 35.36s/it]
 72%|███████▏  | 33/46 [24:11<10:03, 46.39s/it]
 74%|███████▍  | 34/46 [25:00<09:25, 47.11s/it]
 76%|███████▌  | 35/46 [25:26<07:28, 40.79s/it]
 78%|███████▊  | 36/46 [26:37<08:18, 49.89s/it]
 80%|████████  | 37/46 [27:26<07:25, 49.52s/it]
 83%|████████▎ | 38/46 [27:29<04:45, 35.64s/it]
 85%|████████▍ | 39/46 [27:32<03:01, 25.94s/it]
 87%|█████���██▋ | 40/46 [28:44<03:58, 39.83s/it]
 89%|████████▉ | 41/46 [29:34<03:33, 42.69s/it]
 91%|█████████▏| 42/46 [30:00<02:30, 37.68s/it]
 93%|█████████▎| 43/46 [30:25<01:42, 34.05s/it]
 96%|█████████▌| 44/46 [31:37<01:30, 45.21s/it]
 98%|█████████▊| 45/46 [32:03<00:39, 39.42s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity/results_2026-06-08-20:25:23_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_0.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_0 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_0.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_0', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_0.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_0/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_0/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/8 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 12%|█▎        | 1/8 [02:04<14:32, 124.68s/it]
 25%|██▌       | 2/8 [04:09<12:28, 124.81s/it]
 38%|███▊      | 3/8 [06:11<10:17, 123.55s/it]
 50%|█████     | 4/8 [07:20<06:47, 101.94s/it]
 62%|██████▎   | 5/8 [09:18<05:23, 107.77s/it]
 75%|███████▌  | 6/8 [11:16<03:42, 111.32s/it]
 88%|████████▊ | 7/8 [13:16<01:53, 113.99s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_0/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_1.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_1 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_1.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_1', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_1.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_1/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_1/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/7 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 14%|█▍        | 1/7 [02:05<12:35, 125.91s/it]
 29%|██▊       | 2/7 [04:08<10:20, 124.04s/it]
 43%|████▎     | 3/7 [06:10<08:12, 123.16s/it]
 57%|█████▋    | 4/7 [08:11<06:06, 122.10s/it]
 71%|███████▏  | 5/7 [10:11<04:02, 121.28s/it]
 86%|████████▌ | 6/7 [12:10<02:00, 120.60s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_1/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_3.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_3 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_3.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_3', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_3.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_3/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_3/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/7 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 14%|█▍        | 1/7 [02:04<12:27, 124.61s/it]
 29%|██▊       | 2/7 [04:06<10:15, 123.17s/it]
 43%|████▎     | 3/7 [06:08<08:09, 122.36s/it]
 57%|█████▋    | 4/7 [08:08<06:04, 121.66s/it]
 71%|███████▏  | 5/7 [10:07<04:01, 120.78s/it]
 86%|████████▌ | 6/7 [12:08<02:00, 120.60s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_3/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_4.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_4 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_4.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_4', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_4.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_4/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_4/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/7 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 14%|█▍        | 1/7 [02:04<12:29, 124.97s/it]
 29%|██▊       | 2/7 [04:08<10:21, 124.28s/it]
 43%|████▎     | 3/7 [05:53<07:41, 115.39s/it]
 57%|█████▋    | 4/7 [07:54<05:52, 117.41s/it]
 71%|███████▏  | 5/7 [09:38<03:45, 112.78s/it]
 86%|████████▌ | 6/7 [11:20<01:48, 108.98s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_4/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_6.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_6 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_6.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_6', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_6.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_6/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_6/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/7 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 14%|█▍        | 1/7 [02:03<12:23, 123.94s/it]
 29%|██▊       | 2/7 [04:05<10:12, 122.60s/it]
 43%|████▎     | 3/7 [05:33<07:06, 106.55s/it]
 57%|█████▋    | 4/7 [07:30<05:33, 111.00s/it]
 71%|███████▏  | 5/7 [09:29<03:47, 113.70s/it]
 86%|████████▌ | 6/7 [11:27<01:55, 115.30s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_6/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Instance_Preservation_shard_7.log

@@ -0,0 +1,39 @@
+[2026-06-08 21:33:56 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation --dimension Instance_Preservation --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_7 --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_7.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_7', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/full_info_shard_7.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Instance_Preservation', dimension=['Instance_Preservation'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_7/results_2026-06-08-21:33:59_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+[INFO:swift] Successfully registered `/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/Instance_detector/swift/llm/dataset/data/dataset_info.json`.
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_7/results_2026-06-08-21:33:59_full_info.json
+[INFO:swift] Successfully loaded /root/ckpts/vbench2/instance_anomaly_detector/model/args.json.
+[INFO:swift] Loading the model using model_dir: /root/ckpts/huggingface/Qwen/Qwen2.5-VL-3B-Instruct
+[INFO:swift] Setting torch_dtype: torch.bfloat16
+Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.52, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+[INFO:swift] model_kwargs: {'device_map': 'cuda:0'}
+
+[INFO:swift] Setting image_factor: 28. You can adjust this hyperparameter through the environment variable: `IMAGE_FACTOR`.
+[INFO:swift] Setting min_pixels: 3136. You can adjust this hyperparameter through the environment variable: `MIN_PIXELS`.
+[INFO:swift] Using environment variable `MAX_PIXELS`, Setting max_pixels: 1003520.
+[INFO:swift] Setting max_ratio: 200. You can adjust this hyperparameter through the environment variable: `MAX_RATIO`.
+[INFO:swift] Using environment variable `VIDEO_MIN_PIXELS`, Setting video_min_pixels: 100352.
+[INFO:swift] Using environment variable `VIDEO_MAX_PIXELS`, Setting video_max_pixels: 602112.
+[INFO:swift] Using environment variable `VIDEO_TOTAL_PIXELS`, Setting video_total_pixels: 90316800.
+[INFO:swift] Setting frame_factor: 2. You can adjust this hyperparameter through the environment variable: `FRAME_FACTOR`.
+[INFO:swift] Setting fps: 2.0. You can adjust this hyperparameter through the environment variable: `FPS`.
+[INFO:swift] Using environment variable `FPS_MIN_FRAMES`, Setting fps_min_frames: 32.
+[INFO:swift] Using environment variable `FPS_MAX_FRAMES`, Setting fps_max_frames: 50.
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+[INFO:swift] default_system: 'You are a helpful and harmless assistant.'
+[INFO:swift] max_length: 128000
+[INFO:swift] response_prefix: ''
+[INFO:swift] agent_template: hermes
+[INFO:swift] norm_bbox: none
+
  0%|          | 0/7 [00:00<?, ?it/s]qwen-vl-utils using decord to read video.
+
 14%|█▍        | 1/7 [02:05<12:32, 125.35s/it]
 29%|██▊       | 2/7 [04:08<10:20, 124.09s/it]
 43%|████▎     | 3/7 [06:11<08:13, 123.35s/it]
 57%|█████▋    | 4/7 [08:11<06:06, 122.27s/it]
 71%|███████▏  | 5/7 [09:54<03:50, 115.21s/it]
 86%|████████▌ | 6/7 [11:54<01:57, 117.00s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation_shards/2026-06-08-21:29:31/shard_7/results_2026-06-08-21:33:59_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Material.log

@@ -0,0 +1,21 @@
+[2026-06-08 19:56:49 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Material --dimension Material --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Material', dimension=['Material'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material/results_2026-06-08-19:56:51_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material/results_2026-06-08-19:56:51_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 19:57:00,798 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+
  0%|          | 0/50 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  2%|▏         | 1/50 [00:16<13:38, 16.71s/it]
  4%|▍         | 2/50 [00:32<13:00, 16.26s/it]
  6%|▌         | 3/50 [00:48<12:36, 16.09s/it]
  8%|▊         | 4/50 [01:04<12:17, 16.03s/it]
 10%|█         | 5/50 [01:20<12:00, 16.02s/it]
 12%|█▏        | 6/50 [01:36<11:43, 16.00s/it]
 14%|█▍        | 7/50 [01:52<11:27, 15.99s/it]
 16%|█▌        | 8/50 [02:08<11:11, 15.98s/it]
 18%|█▊        | 9/50 [02:24<10:55, 15.99s/it]
 20%|██        | 10/50 [02:40<10:40, 16.00s/it]
 22%|██▏       | 11/50 [02:56<10:22, 15.96s/it]
 24%|██▍       | 12/50 [03:12<10:07, 15.98s/it]
 26%|██▌       | 13/50 [03:28<09:51, 15.98s/it]
 28%|██▊       | 14/50 [03:44<09:34, 15.95s/it]
 30%|███       | 15/50 [03:59<09:15, 15.88s/it]
 32%|███▏      | 16/50 [04:15<08:59, 15.88s/it]
 34%|███▍      | 17/50 [04:31<08:44, 15.91s/it]
 36%|███▌      | 18/50 [04:47<08:26, 15.83s/it]
 38%|███▊      | 19/50 [05:03<08:11, 15.85s/it]
 40%|████      | 20/50 [05:18<07:54, 15.81s/it]
 42%|████▏     | 21/50 [05:34<07:37, 15.77s/it]
 44%|████▍     | 22/50 [05:50<07:20, 15.75s/it]
 46%|████▌     | 23/50 [06:06<07:04, 15.73s/it]
 48%|████▊     | 24/50 [06:21<06:49, 15.75s/it]
 50%|█████     | 25/50 [06:37<06:32, 15.71s/it]
 52%|█████▏    | 26/50 [06:53<06:17, 15.74s/it]
 54%|█████▍    | 27/50 [07:09<06:02, 15.76s/it]
 56%|█████▌    | 28/50 [07:24<05:46, 15.77s/it]
 58%|█████▊    | 29/50 [07:40<05:30, 15.76s/it]
 60%|██████    | 30/50 [07:56<05:14, 15.71s/it]
 62%|██████▏   | 31/50 [08:11<04:58, 15.74s/it]
 64%|██████▍   | 32/50 [08:27<04:43, 15.72s/it]
 66%|██████▌   | 33/50 [08:43<04:27, 15.75s/it]
 68%|██████▊   | 34/50 [08:59<04:12, 15.76s/it]
 70%|███████   | 35/50 [09:14<03:55, 15.72s/it]
 72%|███████▏  | 36/50 [09:30<03:40, 15.76s/it]
 74%|███████▍  | 37/50 [09:46<03:24, 15.76s/it]
 76%|███████▌  | 38/50 [10:02<03:08, 15.75s/it]
 78%|███████▊  | 39/50 [10:17<02:52, 15.70s/it]
 80%|████████  | 40/50 [10:33<02:36, 15.69s/it]
 82%|████████▏ | 41/50 [10:49<02:21, 15.69s/it]
 84%|████████▍ | 42/50 [11:04<02:05, 15.71s/it]
 86%|████████▌ | 43/50 [11:20<01:49, 15.70s/it]
 88%|████████▊ | 44/50 [11:36<01:34, 15.68s/it]
 90%|█████████ | 45/50 [11:45<01:09, 13.83s/it]
 92%|█████████▏| 46/50 [11:54<00:49, 12.45s/it]
 94%|█████████▍| 47/50 [12:04<00:34, 11.48s/it]
 96%|█████████▌| 48/50 [12:13<00:21, 10.84s/it]
 98%|█████████▊| 49/50 [12:22<00:10, 10.38s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material/results_2026-06-08-19:56:51_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Motion_Order_Understanding.log

@@ -0,0 +1,24 @@
+[2026-06-08 18:01:52 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Motion_Order_Understanding --dimension Motion_Order_Understanding --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Motion_Order_Understanding', dimension=['Motion_Order_Understanding'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+File exists: /root/ckpts/vbench2/Qwen/Qwen2.5-7B-Instruct
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding/results_2026-06-08-18:01:56_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding/results_2026-06-08-18:01:56_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 18:02:07,290 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+2026-06-08 18:02:13,735 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+
  0%|          | 0/99 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  1%|          | 1/99 [00:29<48:29, 29.69s/it]
  2%|▏         | 2/99 [00:59<47:53, 29.62s/it]
  3%|▎         | 3/99 [01:26<45:52, 28.68s/it]
  4%|▍         | 4/99 [01:55<45:15, 28.59s/it]
  5%|▌         | 5/99 [02:21<43:13, 27.59s/it]
  6%|▌         | 6/99 [02:51<44:14, 28.55s/it]
  7%|▋         | 7/99 [03:18<42:56, 28.00s/it]
  8%|▊         | 8/99 [03:46<42:37, 28.10s/it]
  9%|▉         | 9/99 [04:12<41:04, 27.39s/it]
 10%|█         | 10/99 [04:40<40:48, 27.51s/it]
 11%|█         | 11/99 [05:14<43:32, 29.69s/it]
 12%|█▏        | 12/99 [05:39<40:59, 28.27s/it]
 13%|█▎        | 13/99 [06:05<39:28, 27.54s/it]
 14%|█▍        | 14/99 [06:32<38:26, 27.13s/it]
 15%|█▌        | 15/99 [06:53<35:39, 25.47s/it]
 16%|█▌        | 16/99 [07:15<33:35, 24.29s/it]
 17%|█▋        | 17/99 [07:34<31:07, 22.77s/it]
 18%|█▊        | 18/99 [07:53<29:14, 21.66s/it]
 19%|█▉        | 19/99 [08:12<27:50, 20.88s/it]
 20%|██        | 20/99 [08:30<26:25, 20.07s/it]
 21%|██        | 21/99 [08:49<25:34, 19.68s/it]
 22%|██▏       | 22/99 [09:08<25:04, 19.54s/it]
 23%|██▎       | 23/99 [09:28<24:46, 19.56s/it]
 24%|██▍       | 24/99 [09:47<24:27, 19.56s/it]
 25%|██▌       | 25/99 [10:08<24:26, 19.82s/it]
 26%|██▋       | 26/99 [10:27<23:52, 19.62s/it]
 27%|██▋       | 27/99 [10:48<23:58, 19.98s/it]
 28%|██▊       | 28/99 [11:06<22:59, 19.44s/it]
 29%|██▉       | 29/99 [11:25<22:29, 19.27s/it]
 30%|███       | 30/99 [11:43<21:53, 19.04s/it]
 31%|███▏      | 31/99 [12:10<24:08, 21.30s/it]
 32%|███▏      | 32/99 [12:40<26:48, 24.01s/it]
 33%|███▎      | 33/99 [13:06<26:52, 24.43s/it]
 34%|███▍      | 34/99 [13:38<28:59, 26.77s/it]
 35%|███▌      | 35/99 [14:05<28:34, 26.79s/it]
 36%|███▋      | 36/99 [14:31<27:50, 26.52s/it]
 37%|███▋      | 37/99 [14:57<27:21, 26.47s/it]
 38%|███▊      | 38/99 [15:25<27:30, 27.05s/it]
 39%|███▉      | 39/99 [15:51<26:43, 26.73s/it]
 40%|████      | 40/99 [16:18<26:17, 26.74s/it]
 41%|████▏     | 41/99 [16:44<25:39, 26.54s/it]
 42%|████▏     | 42/99 [17:13<25:57, 27.32s/it]
 43%|████▎     | 43/99 [17:37<24:29, 26.24s/it]
 44%|████▍     | 44/99 [18:03<23:55, 26.09s/it]
 45%|████▌     | 45/99 [18:29<23:25, 26.03s/it]
 46%|████▋     | 46/99 [18:57<23:39, 26.79s/it]
 47%|████▋     | 47/99 [19:25<23:32, 27.16s/it]
 48%|████▊     | 48/99 [19:46<21:24, 25.18s/it]
 49%|████▉     | 49/99 [20:04<19:21, 23.23s/it]
 51%|█████     | 50/99 [20:24<18:10, 22.25s/it]
 52%|█████▏    | 51/99 [20:45<17:17, 21.62s/it]
 53%|█████▎    | 52/99 [21:05<16:32, 21.12s/it]
 54%|█████▎    | 53/99 [21:23<15:34, 20.32s/it]
 55%|█████▍    | 54/99 [21:41<14:48, 19.74s/it]
 56%|█████▌    | 55/99 [22:02<14:36, 19.93s/it]
 57%|█████▋    | 56/99 [22:23<14:37, 20.40s/it]
 58%|█████▊    | 57/99 [22:42<13:54, 19.87s/it]
 59%|█████▊    | 58/99 [23:02<13:33, 19.84s/it]
 60%|█████▉    | 59/99 [23:21<13:02, 19.57s/it]
 61%|██████    | 60/99 [23:40<12:42, 19.56s/it]
 62%|██████▏   | 61/99 [24:00<12:22, 19.54s/it]
 63%|██████▎   | 62/99 [24:21<12:22, 20.06s/it]
 64%|██████▎   | 63/99 [24:45<12:47, 21.32s/it]
 65%|██████▍   | 64/99 [25:13<13:35, 23.31s/it]
 66%|██████▌   | 65/99 [25:40<13:49, 24.41s/it]
 67%|██████▋   | 66/99 [26:04<13:19, 24.24s/it]
 68%|██████▊   | 67/99 [26:33<13:42, 25.69s/it]
 69%|██████▊   | 68/99 [26:59<13:23, 25.92s/it]
 70%|██████▉   | 69/99 [27:32<13:57, 27.93s/it]
 71%|███████   | 70/99 [27:55<12:49, 26.54s/it]
 72%|███████▏  | 71/99 [28:25<12:50, 27.51s/it]
 73%|███████▎  | 72/99 [28:55<12:45, 28.34s/it]
 74%|███████▎  | 73/99 [29:22<12:00, 27.70s/it]
 75%|███████▍  | 74/99 [29:50<11:38, 27.93s/it]
 76%|███████▌  | 75/99 [30:18<11:13, 28.07s/it]
 77%|███████▋  | 76/99 [30:44<10:24, 27.16s/it]
 78%|███████▊  | 77/99 [31:12<10:03, 27.45s/it]
 79%|███████▉  | 78/99 [31:41<09:49, 28.07s/it]
 80%|███████▉  | 79/99 [32:08<09:13, 27.65s/it]
 81%|████████  | 80/99 [32:28<08:03, 25.43s/it]
 82%|████████▏ | 81/99 [32:47<07:02, 23.49s/it]
 83%|████████▎ | 82/99 [33:08<06:26, 22.73s/it]
 84%|████████▍ | 83/99 [33:27<05:44, 21.55s/it]
 85%|████████▍ | 84/99 [33:46<05:14, 20.94s/it]
 86%|████████▌ | 85/99 [34:05<04:42, 20.15s/it]
 87%|████████▋ | 86/99 [34:23<04:16, 19.71s/it]
 88%|████████▊ | 87/99 [34:45<04:04, 20.40s/it]
 89%|████████▉ | 88/99 [35:04<03:39, 19.98s/it]
 90%|████████▉ | 89/99 [35:24<03:18, 19.86s/it]
 91%|█████████ | 90/99 [35:43<02:57, 19.67s/it]
 92%|█████████▏| 91/99 [36:03<02:37, 19.65s/it]
 93%|█████████▎| 92/99 [36:20<02:12, 18.97s/it]
 94%|█████████▍| 93/99 [36:42<01:58, 19.82s/it]
 95%|█████████▍| 94/99 [37:03<01:40, 20.19s/it]
 96%|█████████▌| 95/99 [37:31<01:29, 22.46s/it]
 97%|█████████▋| 96/99 [37:59<01:12, 24.28s/it]
 98%|█████████▊| 97/99 [38:25<00:49, 24.82s/it]
 99%|█████████▉| 98/99 [38:55<00:26, 26.37s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding/results_2026-06-08-18:01:56_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Multi-View_Consistency.log

@@ -0,0 +1,302 @@
+[2026-06-08 17:28:50 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Multi-View_Consistency --dimension Multi-View_Consistency --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Multi-View_Consistency', dimension=['Multi-View_Consistency'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency/results_2026-06-08-17:28:53_full_info.json
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency/results_2026-06-08-17:28:53_full_info.json
+Using cache found in /root/ckpts/torch/hub/facebookresearch_co-tracker_main
+
  0%|          | 0/97 [00:00<?, ?it/s]2026-06-08 17:28:58,502 - pyscenedetect - INFO - Detecting scenes...
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/torch/functional.py:505: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /pytorch/aten/src/ATen/native/TensorShape.cpp:4317.)
+  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]
+2026-06-08 17:29:06,860 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:10,937 - pyscenedetect - INFO - Detecting scenes...
+
  1%|          | 1/97 [00:17<27:19, 17.07s/it]2026-06-08 17:29:15,578 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:19,600 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:23,563 - pyscenedetect - INFO - Detecting scenes...
+
  2%|▏         | 2/97 [00:29<22:37, 14.29s/it]2026-06-08 17:29:27,922 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:32,229 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:36,479 - pyscenedetect - INFO - Detecting scenes...
+
  3%|▎         | 3/97 [00:42<21:14, 13.56s/it]2026-06-08 17:29:40,606 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:44,604 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:48,669 - pyscenedetect - INFO - Detecting scenes...
+
  4%|▍         | 4/97 [00:54<20:16, 13.08s/it]2026-06-08 17:29:52,949 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:29:57,036 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:01,194 - pyscenedetect - INFO - Detecting scenes...
+
  5%|▌         | 5/97 [01:06<19:36, 12.79s/it]2026-06-08 17:30:05,233 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:09,666 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:13,774 - pyscenedetect - INFO - Detecting scenes...
+
  6%|▌         | 6/97 [01:19<19:18, 12.74s/it]2026-06-08 17:30:17,850 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:21,881 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:25,900 - pyscenedetect - INFO - Detecting scenes...
+
  7%|▋         | 7/97 [01:31<18:43, 12.49s/it]2026-06-08 17:30:29,828 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:33,837 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:44,932 - pyscenedetect - INFO - Detecting scenes...
+
  8%|▊         | 8/97 [01:54<23:46, 16.03s/it]2026-06-08 17:30:53,446 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:30:57,435 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:01,671 - pyscenedetect - INFO - Detecting scenes...
+
  9%|▉         | 9/97 [02:07<21:50, 14.89s/it]2026-06-08 17:31:05,832 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:10,860 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:14,979 - pyscenedetect - INFO - Detecting scenes...
+
 10%|█         | 10/97 [02:20<20:49, 14.36s/it]2026-06-08 17:31:18,996 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:23,160 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:27,152 - pyscenedetect - INFO - Detecting scenes...
+
 11%|█▏        | 11/97 [02:32<19:37, 13.69s/it]2026-06-08 17:31:31,179 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:35,256 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:39,199 - pyscenedetect - INFO - Detecting scenes...
+
 12%|█▏        | 12/97 [02:44<18:44, 13.23s/it]2026-06-08 17:31:43,368 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:53,635 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:31:58,581 - pyscenedetect - INFO - Detecting scenes...
+
 13%|█▎        | 13/97 [03:04<21:16, 15.19s/it]2026-06-08 17:32:03,060 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:32:07,109 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:32:10,987 - pyscenedetect - INFO - Detecting scenes...
+
 14%|█▍        | 14/97 [03:16<19:48, 14.32s/it]2026-06-08 17:32:15,380 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:32:19,671 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:32:23,630 - pyscenedetect - INFO - Detecting scenes...
+
 15%|█▌        | 15/97 [03:29<18:47, 13.75s/it]2026-06-08 17:32:27,786 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:32:31,843 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:32:35,836 - pyscenedetect - INFO - Detecting scenes...
+
 16%|█▋        | 16/97 [03:48<20:36, 15.26s/it]2026-06-08 17:32:46,568 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:32:51,335 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:33:01,035 - pyscenedetect - INFO - Detecting scenes...
+
 18%|█▊        | 17/97 [04:07<21:50, 16.38s/it]2026-06-08 17:33:05,554 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:33:09,641 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:33:20,901 - pyscenedetect - INFO - Detecting scenes...
+
 19%|█▊        | 18/97 [04:31<24:41, 18.76s/it]2026-06-08 17:33:29,862 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:33:39,958 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:33:55,884 - pyscenedetect - INFO - Detecting scenes...
+
 20%|█▉        | 19/97 [05:04<29:57, 23.04s/it]2026-06-08 17:34:02,883 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:34:09,737 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:34:28,947 - pyscenedetect - INFO - Detecting scenes...
+
 21%|██        | 20/97 [05:36<33:15, 25.92s/it]2026-06-08 17:34:35,514 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:34:42,117 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:35:03,701 - pyscenedetect - INFO - Detecting scenes...
+
 22%|██▏       | 21/97 [06:12<36:17, 28.66s/it]2026-06-08 17:35:10,536 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:35:27,169 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:35:44,409 - pyscenedetect - INFO - Detecting scenes...
+
 23%|██▎       | 22/97 [07:06<45:25, 36.34s/it]2026-06-08 17:36:04,787 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:36:11,107 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:36:17,618 - pyscenedetect - INFO - Detecting scenes...
+
 24%|██▎       | 23/97 [07:25<38:33, 31.26s/it]2026-06-08 17:36:24,186 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:36:30,736 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:36:37,427 - pyscenedetect - INFO - Detecting scenes...
+
 25%|██▍       | 24/97 [07:45<33:49, 27.80s/it]2026-06-08 17:36:43,913 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:36:51,060 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:37:09,182 - pyscenedetect - INFO - Detecting scenes...
+
 26%|██▌       | 25/97 [08:29<39:20, 32.78s/it]2026-06-08 17:37:28,346 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:37:34,538 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:37:52,067 - pyscenedetect - INFO - Detecting scenes...
+
 27%|██▋       | 26/97 [09:00<37:59, 32.11s/it]2026-06-08 17:37:58,857 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:38:06,066 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:38:12,755 - pyscenedetect - INFO - Detecting scenes...
+
 28%|██▊       | 27/97 [09:21<33:27, 28.68s/it]2026-06-08 17:38:19,545 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:38:26,054 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:38:32,871 - pyscenedetect - INFO - Detecting scenes...
+
 29%|██▉       | 28/97 [09:41<30:01, 26.11s/it]2026-06-08 17:38:39,664 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:38:46,187 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:39:05,278 - pyscenedetect - INFO - Detecting scenes...
+
 30%|██▉       | 29/97 [10:13<31:49, 28.07s/it]2026-06-08 17:39:12,337 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:39:18,712 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:39:25,067 - pyscenedetect - INFO - Detecting scenes...
+
 31%|███       | 30/97 [10:40<30:52, 27.65s/it]2026-06-08 17:39:38,976 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:39:45,827 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:39:52,407 - pyscenedetect - INFO - Detecting scenes...
+
 32%|███▏      | 31/97 [11:00<27:56, 25.41s/it]2026-06-08 17:39:59,142 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:40:05,960 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:40:12,576 - pyscenedetect - INFO - Detecting scenes...
+
 33%|███▎      | 32/97 [11:20<25:44, 23.76s/it]2026-06-08 17:40:19,059 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:40:25,511 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:40:32,728 - pyscenedetect - INFO - Detecting scenes...
+
 34%|███▍      | 33/97 [11:41<24:22, 22.85s/it]2026-06-08 17:40:39,789 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:40:46,212 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:40:52,976 - pyscenedetect - INFO - Detecting scenes...
+
 35%|███▌      | 34/97 [12:01<23:02, 21.94s/it]2026-06-08 17:40:59,616 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:41:06,431 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:41:25,863 - pyscenedetect - INFO - Detecting scenes...
+
 36%|███▌      | 35/97 [12:34<26:06, 25.26s/it]2026-06-08 17:41:32,638 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:41:47,697 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:41:54,278 - pyscenedetect - INFO - Detecting scenes...
+
 37%|███▋      | 36/97 [13:01<26:27, 26.02s/it]2026-06-08 17:42:00,418 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:42:06,342 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:42:17,872 - pyscenedetect - INFO - Detecting scenes...
+
 38%|███▊      | 37/97 [13:23<24:43, 24.73s/it]2026-06-08 17:42:22,125 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:42:26,150 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:42:30,206 - pyscenedetect - INFO - Detecting scenes...
+
 39%|███▉      | 38/97 [13:36<20:40, 21.03s/it]2026-06-08 17:42:34,528 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:42:43,313 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:42:47,228 - pyscenedetect - INFO - Detecting scenes...
+
 40%|████      | 39/97 [13:57<20:22, 21.08s/it]2026-06-08 17:42:55,729 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:42:59,918 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:43:03,920 - pyscenedetect - INFO - Detecting scenes...
+
 41%|████      | 40/97 [14:09<17:27, 18.38s/it]2026-06-08 17:43:07,812 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:43:12,029 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:43:16,622 - pyscenedetect - INFO - Detecting scenes...
+
 42%|████▏     | 41/97 [14:22<15:42, 16.83s/it]2026-06-08 17:43:21,021 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:43:25,425 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:43:29,521 - pyscenedetect - INFO - Detecting scenes...
+
 43%|████▎     | 42/97 [14:35<14:15, 15.55s/it]2026-06-08 17:43:33,576 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:43:44,665 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:43:48,854 - pyscenedetect - INFO - Detecting scenes...
+
 44%|████▍     | 43/97 [14:54<15:02, 16.72s/it]2026-06-08 17:43:53,029 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:01,037 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:07,835 - pyscenedetect - INFO - Detecting scenes...
+
 45%|████▌     | 44/97 [15:13<15:21, 17.39s/it]2026-06-08 17:44:11,995 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:16,610 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:20,624 - pyscenedetect - INFO - Detecting scenes...
+
 46%|████▋     | 45/97 [15:26<13:51, 15.99s/it]2026-06-08 17:44:24,702 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:28,955 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:33,017 - pyscenedetect - INFO - Detecting scenes...
+
 47%|████▋     | 46/97 [15:38<12:37, 14.85s/it]2026-06-08 17:44:36,894 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:46,490 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:44:55,064 - pyscenedetect - INFO - Detecting scenes...
+
 48%|████▊     | 47/97 [16:00<14:15, 17.11s/it]2026-06-08 17:44:59,272 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:03,228 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:13,432 - pyscenedetect - INFO - Detecting scenes...
+
 49%|████▉     | 48/97 [16:23<15:21, 18.81s/it]2026-06-08 17:45:22,061 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:26,228 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:30,387 - pyscenedetect - INFO - Detecting scenes...
+
 51%|█████     | 49/97 [16:35<13:30, 16.88s/it]2026-06-08 17:45:34,422 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:38,256 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:43,071 - pyscenedetect - INFO - Detecting scenes...
+
 52%|█████▏    | 50/97 [16:48<12:15, 15.65s/it]2026-06-08 17:45:47,225 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:51,265 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:45:55,415 - pyscenedetect - INFO - Detecting scenes...
+
 53%|█████▎    | 51/97 [17:00<11:12, 14.61s/it]2026-06-08 17:45:59,421 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:46:03,626 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:46:07,877 - pyscenedetect - INFO - Detecting scenes...
+
 54%|█████▎    | 52/97 [17:14<10:42, 14.29s/it]2026-06-08 17:46:12,943 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:46:21,848 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:46:25,827 - pyscenedetect - INFO - Detecting scenes...
+
 55%|█████▍    | 53/97 [17:31<11:01, 15.03s/it]2026-06-08 17:46:29,702 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:46:33,991 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:46:38,350 - pyscenedetect - INFO - Detecting scenes...
+
 56%|█████▌    | 54/97 [17:54<12:38, 17.64s/it]2026-06-08 17:46:53,415 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:46:57,650 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:47:01,719 - pyscenedetect - INFO - Detecting scenes...
+
 57%|█████▋    | 55/97 [18:07<11:14, 16.06s/it]2026-06-08 17:47:05,791 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:47:10,294 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:47:14,390 - pyscenedetect - INFO - Detecting scenes...
+
 58%|█████▊    | 56/97 [18:19<10:16, 15.04s/it]2026-06-08 17:47:18,441 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:47:23,150 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:47:41,347 - pyscenedetect - INFO - Detecting scenes...
+
 59%|█████▉    | 57/97 [18:59<14:51, 22.29s/it]2026-06-08 17:47:57,656 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:48:03,722 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:48:10,318 - pyscenedetect - INFO - Detecting scenes...
+
 60%|█████▉    | 58/97 [19:18<13:54, 21.41s/it]2026-06-08 17:48:17,256 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:48:35,200 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:48:42,079 - pyscenedetect - INFO - Detecting scenes...
+
 61%|██████    | 59/97 [19:50<15:37, 24.66s/it]2026-06-08 17:48:49,247 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:49:06,611 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:49:12,927 - pyscenedetect - INFO - Detecting scenes...
+
 62%|██████▏   | 60/97 [20:20<16:14, 26.34s/it]2026-06-08 17:49:19,512 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:49:25,597 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:49:43,796 - pyscenedetect - INFO - Detecting scenes...
+
 63%|██████▎   | 61/97 [20:52<16:41, 27.83s/it]2026-06-08 17:49:50,816 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:49:57,851 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:50:04,404 - pyscenedetect - INFO - Detecting scenes...
+
 64%|██████▍   | 62/97 [21:12<14:55, 25.59s/it]2026-06-08 17:50:11,184 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:50:17,652 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:50:23,921 - pyscenedetect - INFO - Detecting scenes...
+
 65%|██████▍   | 63/97 [21:32<13:26, 23.73s/it]2026-06-08 17:50:30,553 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:50:48,763 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:50:55,116 - pyscenedetect - INFO - Detecting scenes...
+
 66%|██████▌   | 64/97 [22:02<14:13, 25.86s/it]2026-06-08 17:51:01,412 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:51:08,294 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:51:14,725 - pyscenedetect - INFO - Detecting scenes...
+
 67%|██████▋   | 65/97 [22:22<12:46, 23.95s/it]2026-06-08 17:51:20,903 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:51:27,973 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:51:35,629 - pyscenedetect - INFO - Detecting scenes...
+
 68%|██████▊   | 66/97 [22:43<11:58, 23.18s/it]2026-06-08 17:51:42,305 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:51:48,562 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:52:07,522 - pyscenedetect - INFO - Detecting scenes...
+
 69%|██████▉   | 67/97 [23:15<12:49, 25.64s/it]2026-06-08 17:52:13,676 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:52:20,000 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:52:26,497 - pyscenedetect - INFO - Detecting scenes...
+
 70%|███████   | 68/97 [23:44<12:58, 26.86s/it]2026-06-08 17:52:43,360 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:52:49,781 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:52:56,543 - pyscenedetect - INFO - Detecting scenes...
+
 71%|███████   | 69/97 [24:04<11:30, 24.67s/it]2026-06-08 17:53:02,927 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:53:09,321 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:53:15,344 - pyscenedetect - INFO - Detecting scenes...
+
 72%|███████▏  | 70/97 [24:23<10:19, 22.93s/it]2026-06-08 17:53:21,777 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:53:28,577 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:53:35,479 - pyscenedetect - INFO - Detecting scenes...
+
 73%|███████▎  | 71/97 [24:53<10:50, 25.02s/it]2026-06-08 17:53:51,683 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:53:58,533 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:54:05,682 - pyscenedetect - INFO - Detecting scenes...
+
 74%|███████▍  | 72/97 [25:14<09:54, 23.78s/it]2026-06-08 17:54:12,575 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:54:18,667 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:54:24,987 - pyscenedetect - INFO - Detecting scenes...
+
 75%|███████▌  | 73/97 [25:33<08:55, 22.33s/it]2026-06-08 17:54:31,513 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:54:38,084 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:54:55,918 - pyscenedetect - INFO - Detecting scenes...
+
 76%|███████▋  | 74/97 [26:04<09:33, 24.94s/it]2026-06-08 17:55:02,545 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:55:19,720 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:55:27,120 - pyscenedetect - INFO - Detecting scenes...
+
 77%|███████▋  | 75/97 [26:34<09:47, 26.71s/it]2026-06-08 17:55:33,435 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:55:39,297 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:55:43,387 - pyscenedetect - INFO - Detecting scenes...
+
 78%|███████▊  | 76/97 [26:49<08:02, 22.98s/it]2026-06-08 17:55:47,637 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:55:51,976 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:55:56,573 - pyscenedetect - INFO - Detecting scenes...
+
 79%|███████▉  | 77/97 [27:02<06:40, 20.01s/it]2026-06-08 17:56:00,720 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:56:04,832 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:56:18,328 - pyscenedetect - INFO - Detecting scenes...
+
 80%|████████  | 78/97 [27:24<06:30, 20.54s/it]2026-06-08 17:56:22,551 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:56:26,547 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:56:40,387 - pyscenedetect - INFO - Detecting scenes...
+
 81%|████████▏ | 79/97 [27:46<06:20, 21.15s/it]2026-06-08 17:56:45,083 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:56:49,220 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:56:53,275 - pyscenedetect - INFO - Detecting scenes...
+
 82%|████████▏ | 80/97 [28:04<05:41, 20.06s/it]2026-06-08 17:57:02,602 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:57:06,677 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:57:10,838 - pyscenedetect - INFO - Detecting scenes...
+
 84%|████████▎ | 81/97 [28:16<04:45, 17.84s/it]2026-06-08 17:57:15,241 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:57:23,498 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:57:35,333 - pyscenedetect - INFO - Detecting scenes...
+
 85%|████████▍ | 82/97 [28:41<04:57, 19.83s/it]2026-06-08 17:57:39,738 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:57:43,891 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:57:47,996 - pyscenedetect - INFO - Detecting scenes...
+
 86%|████████▌ | 83/97 [28:53<04:05, 17.55s/it]2026-06-08 17:57:51,977 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:57:55,924 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:00,040 - pyscenedetect - INFO - Detecting scenes...
+
 87%|████████▋ | 84/97 [29:05<03:27, 15.93s/it]2026-06-08 17:58:04,125 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:12,657 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:16,670 - pyscenedetect - INFO - Detecting scenes...
+
 88%|████████▊ | 85/97 [29:22<03:13, 16.14s/it]2026-06-08 17:58:20,754 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:24,848 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:28,903 - pyscenedetect - INFO - Detecting scenes...
+
 89%|████████▊ | 86/97 [29:34<02:45, 15.00s/it]2026-06-08 17:58:33,094 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:37,145 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:41,169 - pyscenedetect - INFO - Detecting scenes...
+
 90%|████████▉ | 87/97 [29:46<02:21, 14.20s/it]2026-06-08 17:58:45,414 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:49,246 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:58:53,384 - pyscenedetect - INFO - Detecting scenes...
+
 91%|█████████ | 88/97 [29:59<02:02, 13.61s/it]2026-06-08 17:58:57,645 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:01,761 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:05,799 - pyscenedetect - INFO - Detecting scenes...
+
 92%|█████████▏| 89/97 [30:11<01:45, 13.23s/it]2026-06-08 17:59:10,023 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:14,368 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:18,248 - pyscenedetect - INFO - Detecting scenes...
+
 93%|█████████▎| 90/97 [30:23<01:30, 12.92s/it]2026-06-08 17:59:22,206 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:26,395 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:31,142 - pyscenedetect - INFO - Detecting scenes...
+
 94%|█████████▍| 91/97 [30:41<01:26, 14.41s/it]2026-06-08 17:59:40,094 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:43,965 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:48,099 - pyscenedetect - INFO - Detecting scenes...
+
 95%|█████████▍| 92/97 [30:54<01:09, 13.97s/it]2026-06-08 17:59:53,031 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 17:59:57,591 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:01,743 - pyscenedetect - INFO - Detecting scenes...
+
 96%|█████████▌| 93/97 [31:07<00:54, 13.61s/it]2026-06-08 18:00:05,795 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:10,028 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:14,441 - pyscenedetect - INFO - Detecting scenes...
+
 97%|█████████▋| 94/97 [31:19<00:39, 13.28s/it]2026-06-08 18:00:18,293 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:22,214 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:26,266 - pyscenedetect - INFO - Detecting scenes...
+
 98%|█████████▊| 95/97 [31:31<00:25, 12.88s/it]2026-06-08 18:00:30,241 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:35,107 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:39,047 - pyscenedetect - INFO - Detecting scenes...
+
 99%|█████████▉| 96/97 [31:44<00:12, 12.88s/it]2026-06-08 18:00:43,119 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:47,233 - pyscenedetect - INFO - Detecting scenes...
+2026-06-08 18:00:51,334 - pyscenedetect - INFO - Detecting scenes...
+
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency/results_2026-06-08-17:28:53_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/Thermotics.log

@@ -0,0 +1,21 @@
+[2026-06-08 20:10:19 UTC] COMMAND /root/ckpts/micromamba/envs/vbench2/bin/python -u evaluate.py --videos_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Thermotics --dimension Thermotics --output_path /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics --full_json_dir /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json --mode vbench_standard
+args: Namespace(output_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics', full_json_dir='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_selected_info.json', videos_path='/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/videos/Thermotics', dimension=['Thermotics'], load_ckpt_from_local=None, read_frame=None, mode='vbench_standard', prompt='None', prompt_file=None, category=None, imaging_quality_preprocessing_mode='longer')
+start evaluation
+File exists: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+Evaluation meta data saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics/results_2026-06-08-20:10:22_full_info.json
+/root/ckpts/micromamba/envs/vbench2/lib/python3.10/site-packages/transformers/utils/hub.py:105: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
+  warnings.warn(
+/nfs/ywang29/Reward_finetuning/VideoX-Fun/VBench/VBench-2.0/vbench2/third_party/LLaVA_NeXT/llava/model/llava_arch.py:215: SyntaxWarning: "is not" with a literal. Did you mean "!="?
+  if slower_img_feat is not 0:
+cur_full_info_path: /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics/results_2026-06-08-20:10:22_full_info.json
+Loaded LLaVA model: /root/ckpts/vbench2/lmms-lab/LLaVA-Video-7B-Qwen2
+You are using a model of type llava to instantiate a model of type llava_qwen. This is not supported for all configurations of models and can yield errors.
+Sliding Window Attention is enabled but not implemented for `sdpa`; unexpected results may be encountered.
+Loading vision tower: google/siglip-so400m-patch14-384
+2026-06-08 20:10:28,748 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
+
+Model Class: LlavaQwenForCausalLM
+
  0%|          | 0/50 [00:00<?, ?it/s]The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+
  2%|▏         | 1/50 [00:16<13:06, 16.05s/it]
  4%|▍         | 2/50 [00:31<12:40, 15.84s/it]
  6%|▌         | 3/50 [00:47<12:19, 15.74s/it]
  8%|▊         | 4/50 [01:02<11:59, 15.64s/it]
 10%|█         | 5/50 [01:18<11:42, 15.60s/it]
 12%|█▏        | 6/50 [01:33<11:25, 15.58s/it]
 14%|█▍        | 7/50 [01:49<11:09, 15.56s/it]
 16%|█▌        | 8/50 [02:05<10:53, 15.56s/it]
 18%|█▊        | 9/50 [02:20<10:37, 15.55s/it]
 20%|██        | 10/50 [02:36<10:21, 15.54s/it]
 22%|██▏       | 11/50 [02:45<08:49, 13.58s/it]
 24%|██▍       | 12/50 [03:00<08:59, 14.21s/it]
 26%|██▌       | 13/50 [03:16<09:01, 14.63s/it]
 28%|██▊       | 14/50 [03:32<08:57, 14.94s/it]
 30%|███       | 15/50 [03:47<08:49, 15.13s/it]
 32%|███▏      | 16/50 [04:03<08:39, 15.28s/it]
 34%|███▍      | 17/50 [04:18<08:27, 15.37s/it]
 36%|███▌      | 18/50 [04:34<08:14, 15.46s/it]
 38%|███▊      | 19/50 [04:50<08:00, 15.49s/it]
 40%|████      | 20/50 [04:59<06:46, 13.54s/it]
 42%|████▏     | 21/50 [05:14<06:50, 14.17s/it]
 44%|████▍     | 22/50 [05:30<06:48, 14.58s/it]
 46%|████▌     | 23/50 [05:39<05:48, 12.92s/it]
 48%|████▊     | 24/50 [05:54<05:56, 13.73s/it]
 50%|���████     | 25/50 [06:10<05:58, 14.33s/it]
 52%|█████▏    | 26/50 [06:26<05:53, 14.72s/it]
 54%|█████▍    | 27/50 [06:41<05:44, 14.98s/it]
 56%|█████▌    | 28/50 [06:57<05:33, 15.17s/it]
 58%|█████▊    | 29/50 [07:13<05:20, 15.28s/it]
 60%|██████    | 30/50 [07:28<05:07, 15.37s/it]
 62%|██████▏   | 31/50 [07:44<04:53, 15.44s/it]
 64%|██████▍   | 32/50 [07:59<04:38, 15.48s/it]
 66%|██████▌   | 33/50 [08:15<04:23, 15.51s/it]
 68%|██████▊   | 34/50 [08:30<04:08, 15.52s/it]
 70%|███████   | 35/50 [08:46<03:53, 15.55s/it]
 72%|███████▏  | 36/50 [09:02<03:37, 15.56s/it]
 74%|███████▍  | 37/50 [09:17<03:22, 15.57s/it]
 76%|███████▌  | 38/50 [09:39<03:30, 17.52s/it]
 78%|███████▊  | 39/50 [09:55<03:06, 16.92s/it]
 80%|████████  | 40/50 [10:10<02:45, 16.51s/it]
 82%|████████▏ | 41/50 [10:26<02:26, 16.23s/it]
 84%|████████▍ | 42/50 [10:42<02:08, 16.04s/it]
 86%|████████▌ | 43/50 [10:51<01:37, 13.95s/it]
 88%|████████▊ | 44/50 [11:00<01:14, 12.48s/it]
 90%|█████████ | 45/50 [11:15<01:07, 13.42s/it]
 92%|█████████▏| 46/50 [11:31<00:56, 14.07s/it]
 94%|█████████▍| 47/50 [11:46<00:43, 14.53s/it]
 96%|█████████▌| 48/50 [12:02<00:29, 14.85s/it]
 98%|█████████▊| 49/50 [12:18<00:15, 15.10s/it]
+Evaluation results saved to /nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics/results_2026-06-08-20:10:22_eval_results.json
+done

outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/logs/vbench2_parallel/parallel_status.json

@@ -0,0 +1,27 @@
+{
+  "generated_at_utc": "2026-06-08T22:11:20.952866+00:00",
+  "run_id": "yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting",
+  "step": "100",
+  "output_root": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results",
+  "completed": {
+    "Camera_Motion": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Camera_Motion/results_2026-06-08-18:42:29_eval_results.json",
+    "Complex_Plot": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Plot/results_2026-06-08-19:23:02_eval_results.json",
+    "Human_Anatomy": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Anatomy/results_2026-06-08-17:28:53_eval_results.json",
+    "Human_Identity": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Identity/results_2026-06-08-20:25:23_eval_results.json",
+    "Human_Clothes": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Clothes/results_2026-06-08-20:23:53_eval_results.json",
+    "Diversity": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Diversity/results_2026-06-08-17:28:53_eval_results.json",
+    "Composition": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Composition/results_2026-06-08-20:51:55_eval_results.json",
+    "Dynamic_Spatial_Relationship": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Spatial_Relationship/results_2026-06-08-19:14:29_eval_results.json",
+    "Dynamic_Attribute": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Dynamic_Attribute/results_2026-06-08-18:53:30_eval_results.json",
+    "Motion_Order_Understanding": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Order_Understanding/results_2026-06-08-18:01:56_eval_results.json",
+    "Human_Interaction": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Human_Interaction/results_2026-06-08-17:28:53_eval_results.json",
+    "Complex_Landscape": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Complex_Landscape/results_2026-06-08-20:58:55_eval_results.json",
+    "Motion_Rationality": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Motion_Rationality/results_2026-06-08-19:56:51_eval_results.json",
+    "Instance_Preservation": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Instance_Preservation/results_2026-06-08-21:49:42_eval_results.json",
+    "Mechanics": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Mechanics/results_2026-06-08-19:23:32_eval_results.json",
+    "Thermotics": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Thermotics/results_2026-06-08-20:10:22_eval_results.json",
+    "Material": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Material/results_2026-06-08-19:56:51_eval_results.json",
+    "Multi-View_Consistency": "/nfs/ywang29/Reward_finetuning/SAGE-GRPO/outputs/yw008-2-sage-videoscore2-token-credit-video-adv-no-weighting/vbench2_eval/step_100/vbench2_results/Multi-View_Consistency/results_2026-06-08-17:28:53_eval_results.json"
+  },
+  "failed": {}
+}