Upload . with huggingface_hub

README.md

@@ -15,7 +15,7 @@ model-index:
       type: doom_health_gathering_supreme
     metrics:
     - type: mean_reward
-      value: 8.10 +/- 3.73
+      value: 9.48 +/- 4.69
       name: mean_reward
       verified: false
 ---

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85544d48b86c629f4a0a237f1d95ca2e788ff0fd8f91aaaca5e4ce0b98a4a76c
-size 14721558
+oid sha256:1b1808132fd6f5e6864ab3695bf2ec741b1bb2be819e19c7e32c48a86b1d0593
+size 17548925

sf_log.txt

@@ -1283,3 +1283,285 @@ main_loop: 1180.8338
 [2023-03-11 11:45:53,574][00127] Avg episode rewards: #0: 17.000, true rewards: #0: 8.100
 [2023-03-11 11:45:53,575][00127] Avg episode reward: 17.000, avg true_objective: 8.100
 [2023-03-11 11:46:41,540][00127] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
+[2023-03-11 11:46:53,587][00127] The model has been pushed to https://huggingface.co/Taratata/rl_course_vizdoom_health_gathering_supreme
+[2023-03-11 11:48:24,968][00127] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2023-03-11 11:48:24,970][00127] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-03-11 11:48:24,972][00127] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-03-11 11:48:24,974][00127] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-03-11 11:48:24,976][00127] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-03-11 11:48:24,977][00127] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-03-11 11:48:24,979][00127] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file!
+[2023-03-11 11:48:24,981][00127] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-03-11 11:48:24,982][00127] Adding new argument 'push_to_hub'=False that is not in the saved config file!
+[2023-03-11 11:48:24,983][00127] Adding new argument 'hf_repository'=None that is not in the saved config file!
+[2023-03-11 11:48:24,984][00127] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-03-11 11:48:24,985][00127] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-03-11 11:48:24,986][00127] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-03-11 11:48:24,987][00127] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-03-11 11:48:24,988][00127] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-03-11 11:48:25,005][00127] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-11 11:48:25,008][00127] RunningMeanStd input shape: (1,)
+[2023-03-11 11:48:25,022][00127] ConvEncoder: input_channels=3
+[2023-03-11 11:48:25,062][00127] Conv encoder output size: 512
+[2023-03-11 11:48:25,063][00127] Policy head output size: 512
+[2023-03-11 11:48:25,082][00127] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
+[2023-03-11 11:48:25,531][00127] Num frames 100...
+[2023-03-11 11:48:25,643][00127] Num frames 200...
+[2023-03-11 11:48:25,755][00127] Num frames 300...
+[2023-03-11 11:48:25,864][00127] Num frames 400...
+[2023-03-11 11:48:25,984][00127] Num frames 500...
+[2023-03-11 11:48:26,098][00127] Num frames 600...
+[2023-03-11 11:48:26,204][00127] Num frames 700...
+[2023-03-11 11:48:26,316][00127] Num frames 800...
+[2023-03-11 11:48:26,435][00127] Num frames 900...
+[2023-03-11 11:48:26,552][00127] Num frames 1000...
+[2023-03-11 11:48:26,668][00127] Avg episode rewards: #0: 22.550, true rewards: #0: 10.550
+[2023-03-11 11:48:26,671][00127] Avg episode reward: 22.550, avg true_objective: 10.550
+[2023-03-11 11:48:26,724][00127] Num frames 1100...
+[2023-03-11 11:48:26,841][00127] Num frames 1200...
+[2023-03-11 11:48:26,952][00127] Num frames 1300...
+[2023-03-11 11:48:27,062][00127] Num frames 1400...
+[2023-03-11 11:48:27,171][00127] Num frames 1500...
+[2023-03-11 11:48:27,288][00127] Num frames 1600...
+[2023-03-11 11:48:27,410][00127] Num frames 1700...
+[2023-03-11 11:48:27,525][00127] Num frames 1800...
+[2023-03-11 11:48:27,645][00127] Num frames 1900...
+[2023-03-11 11:48:27,754][00127] Num frames 2000...
+[2023-03-11 11:48:27,864][00127] Num frames 2100...
+[2023-03-11 11:48:27,986][00127] Num frames 2200...
+[2023-03-11 11:48:28,109][00127] Num frames 2300...
+[2023-03-11 11:48:28,220][00127] Avg episode rewards: #0: 27.245, true rewards: #0: 11.745
+[2023-03-11 11:48:28,222][00127] Avg episode reward: 27.245, avg true_objective: 11.745
+[2023-03-11 11:48:28,284][00127] Num frames 2400...
+[2023-03-11 11:48:28,395][00127] Num frames 2500...
+[2023-03-11 11:48:28,507][00127] Num frames 2600...
+[2023-03-11 11:48:28,624][00127] Num frames 2700...
+[2023-03-11 11:48:28,747][00127] Num frames 2800...
+[2023-03-11 11:48:28,858][00127] Num frames 2900...
+[2023-03-11 11:48:28,968][00127] Num frames 3000...
+[2023-03-11 11:48:29,074][00127] Num frames 3100...
+[2023-03-11 11:48:29,186][00127] Num frames 3200...
+[2023-03-11 11:48:29,301][00127] Num frames 3300...
+[2023-03-11 11:48:29,425][00127] Num frames 3400...
+[2023-03-11 11:48:29,535][00127] Num frames 3500...
+[2023-03-11 11:48:29,652][00127] Num frames 3600...
+[2023-03-11 11:48:29,764][00127] Num frames 3700...
+[2023-03-11 11:48:29,879][00127] Num frames 3800...
+[2023-03-11 11:48:29,989][00127] Num frames 3900...
+[2023-03-11 11:48:30,118][00127] Num frames 4000...
+[2023-03-11 11:48:30,231][00127] Num frames 4100...
+[2023-03-11 11:48:30,340][00127] Num frames 4200...
+[2023-03-11 11:48:30,472][00127] Avg episode rewards: #0: 35.563, true rewards: #0: 14.230
+[2023-03-11 11:48:30,474][00127] Avg episode reward: 35.563, avg true_objective: 14.230
+[2023-03-11 11:48:30,517][00127] Num frames 4300...
+[2023-03-11 11:48:30,634][00127] Num frames 4400...
+[2023-03-11 11:48:30,753][00127] Num frames 4500...
+[2023-03-11 11:48:30,864][00127] Num frames 4600...
+[2023-03-11 11:48:30,978][00127] Num frames 4700...
+[2023-03-11 11:48:31,088][00127] Num frames 4800...
+[2023-03-11 11:48:31,199][00127] Num frames 4900...
+[2023-03-11 11:48:31,310][00127] Num frames 5000...
+[2023-03-11 11:48:31,408][00127] Avg episode rewards: #0: 30.592, true rewards: #0: 12.592
+[2023-03-11 11:48:31,410][00127] Avg episode reward: 30.592, avg true_objective: 12.592
+[2023-03-11 11:48:31,481][00127] Num frames 5100...
+[2023-03-11 11:48:31,590][00127] Num frames 5200...
+[2023-03-11 11:48:31,710][00127] Num frames 5300...
+[2023-03-11 11:48:31,822][00127] Num frames 5400...
+[2023-03-11 11:48:31,931][00127] Num frames 5500...
+[2023-03-11 11:48:32,062][00127] Num frames 5600...
+[2023-03-11 11:48:32,169][00127] Num frames 5700...
+[2023-03-11 11:48:32,284][00127] Num frames 5800...
+[2023-03-11 11:48:32,398][00127] Num frames 5900...
+[2023-03-11 11:48:32,513][00127] Num frames 6000...
+[2023-03-11 11:48:32,621][00127] Num frames 6100...
+[2023-03-11 11:48:32,743][00127] Num frames 6200...
+[2023-03-11 11:48:32,862][00127] Num frames 6300...
+[2023-03-11 11:48:32,959][00127] Avg episode rewards: #0: 30.476, true rewards: #0: 12.676
+[2023-03-11 11:48:32,960][00127] Avg episode reward: 30.476, avg true_objective: 12.676
+[2023-03-11 11:48:33,029][00127] Num frames 6400...
+[2023-03-11 11:48:33,138][00127] Num frames 6500...
+[2023-03-11 11:48:33,295][00127] Num frames 6600...
+[2023-03-11 11:48:33,485][00127] Avg episode rewards: #0: 26.150, true rewards: #0: 11.150
+[2023-03-11 11:48:33,487][00127] Avg episode reward: 26.150, avg true_objective: 11.150
+[2023-03-11 11:48:33,516][00127] Num frames 6700...
+[2023-03-11 11:48:33,686][00127] Num frames 6800...
+[2023-03-11 11:48:33,842][00127] Num frames 6900...
+[2023-03-11 11:48:33,997][00127] Num frames 7000...
+[2023-03-11 11:48:34,151][00127] Num frames 7100...
+[2023-03-11 11:48:34,303][00127] Num frames 7200...
+[2023-03-11 11:48:34,460][00127] Num frames 7300...
+[2023-03-11 11:48:34,612][00127] Num frames 7400...
+[2023-03-11 11:48:34,786][00127] Num frames 7500...
+[2023-03-11 11:48:34,936][00127] Num frames 7600...
+[2023-03-11 11:48:35,092][00127] Num frames 7700...
+[2023-03-11 11:48:35,266][00127] Avg episode rewards: #0: 26.254, true rewards: #0: 11.111
+[2023-03-11 11:48:35,268][00127] Avg episode reward: 26.254, avg true_objective: 11.111
+[2023-03-11 11:48:35,311][00127] Num frames 7800...
+[2023-03-11 11:48:35,470][00127] Num frames 7900...
+[2023-03-11 11:48:35,629][00127] Num frames 8000...
+[2023-03-11 11:48:35,788][00127] Num frames 8100...
+[2023-03-11 11:48:35,899][00127] Avg episode rewards: #0: 23.542, true rewards: #0: 10.167
+[2023-03-11 11:48:35,901][00127] Avg episode reward: 23.542, avg true_objective: 10.167
+[2023-03-11 11:48:36,004][00127] Num frames 8200...
+[2023-03-11 11:48:36,157][00127] Num frames 8300...
+[2023-03-11 11:48:36,312][00127] Num frames 8400...
+[2023-03-11 11:48:36,466][00127] Num frames 8500...
+[2023-03-11 11:48:36,632][00127] Num frames 8600...
+[2023-03-11 11:48:36,761][00127] Num frames 8700...
+[2023-03-11 11:48:36,879][00127] Num frames 8800...
+[2023-03-11 11:48:36,997][00127] Num frames 8900...
+[2023-03-11 11:48:37,115][00127] Num frames 9000...
+[2023-03-11 11:48:37,223][00127] Num frames 9100...
+[2023-03-11 11:48:37,335][00127] Num frames 9200...
+[2023-03-11 11:48:37,492][00127] Avg episode rewards: #0: 23.762, true rewards: #0: 10.318
+[2023-03-11 11:48:37,494][00127] Avg episode reward: 23.762, avg true_objective: 10.318
+[2023-03-11 11:48:37,514][00127] Num frames 9300...
+[2023-03-11 11:48:37,624][00127] Num frames 9400...
+[2023-03-11 11:48:37,732][00127] Num frames 9500...
+[2023-03-11 11:48:37,849][00127] Num frames 9600...
+[2023-03-11 11:48:37,962][00127] Num frames 9700...
+[2023-03-11 11:48:38,073][00127] Num frames 9800...
+[2023-03-11 11:48:38,183][00127] Num frames 9900...
+[2023-03-11 11:48:38,301][00127] Num frames 10000...
+[2023-03-11 11:48:38,411][00127] Num frames 10100...
+[2023-03-11 11:48:38,520][00127] Num frames 10200...
+[2023-03-11 11:48:38,627][00127] Num frames 10300...
+[2023-03-11 11:48:38,732][00127] Avg episode rewards: #0: 23.841, true rewards: #0: 10.341
+[2023-03-11 11:48:38,733][00127] Avg episode reward: 23.841, avg true_objective: 10.341
+[2023-03-11 11:49:39,627][00127] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
+[2023-03-11 11:50:02,077][00127] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2023-03-11 11:50:02,079][00127] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-03-11 11:50:02,081][00127] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-03-11 11:50:02,083][00127] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-03-11 11:50:02,085][00127] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-03-11 11:50:02,087][00127] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-03-11 11:50:02,088][00127] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
+[2023-03-11 11:50:02,090][00127] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-03-11 11:50:02,091][00127] Adding new argument 'push_to_hub'=True that is not in the saved config file!
+[2023-03-11 11:50:02,092][00127] Adding new argument 'hf_repository'='Taratata/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
+[2023-03-11 11:50:02,093][00127] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-03-11 11:50:02,094][00127] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-03-11 11:50:02,095][00127] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-03-11 11:50:02,096][00127] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-03-11 11:50:02,097][00127] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-03-11 11:50:02,117][00127] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-11 11:50:02,120][00127] RunningMeanStd input shape: (1,)
+[2023-03-11 11:50:02,133][00127] ConvEncoder: input_channels=3
+[2023-03-11 11:50:02,168][00127] Conv encoder output size: 512
+[2023-03-11 11:50:02,172][00127] Policy head output size: 512
+[2023-03-11 11:50:02,192][00127] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
+[2023-03-11 11:50:02,628][00127] Num frames 100...
+[2023-03-11 11:50:02,748][00127] Num frames 200...
+[2023-03-11 11:50:02,868][00127] Num frames 300...
+[2023-03-11 11:50:02,990][00127] Num frames 400...
+[2023-03-11 11:50:03,108][00127] Num frames 500...
+[2023-03-11 11:50:03,227][00127] Num frames 600...
+[2023-03-11 11:50:03,340][00127] Num frames 700...
+[2023-03-11 11:50:03,459][00127] Num frames 800...
+[2023-03-11 11:50:03,580][00127] Num frames 900...
+[2023-03-11 11:50:03,696][00127] Num frames 1000...
+[2023-03-11 11:50:03,768][00127] Avg episode rewards: #0: 22.080, true rewards: #0: 10.080
+[2023-03-11 11:50:03,770][00127] Avg episode reward: 22.080, avg true_objective: 10.080
+[2023-03-11 11:50:03,875][00127] Num frames 1100...
+[2023-03-11 11:50:03,999][00127] Num frames 1200...
+[2023-03-11 11:50:04,108][00127] Num frames 1300...
+[2023-03-11 11:50:04,224][00127] Num frames 1400...
+[2023-03-11 11:50:04,339][00127] Num frames 1500...
+[2023-03-11 11:50:04,450][00127] Num frames 1600...
+[2023-03-11 11:50:04,566][00127] Num frames 1700...
+[2023-03-11 11:50:04,685][00127] Num frames 1800...
+[2023-03-11 11:50:04,801][00127] Num frames 1900...
+[2023-03-11 11:50:04,909][00127] Num frames 2000...
+[2023-03-11 11:50:05,022][00127] Num frames 2100...
+[2023-03-11 11:50:05,132][00127] Num frames 2200...
+[2023-03-11 11:50:05,247][00127] Num frames 2300...
+[2023-03-11 11:50:05,357][00127] Num frames 2400...
+[2023-03-11 11:50:05,469][00127] Num frames 2500...
+[2023-03-11 11:50:05,580][00127] Num frames 2600...
+[2023-03-11 11:50:05,695][00127] Num frames 2700...
+[2023-03-11 11:50:05,807][00127] Num frames 2800...
+[2023-03-11 11:50:05,859][00127] Avg episode rewards: #0: 32.500, true rewards: #0: 14.000
+[2023-03-11 11:50:05,861][00127] Avg episode reward: 32.500, avg true_objective: 14.000
+[2023-03-11 11:50:05,977][00127] Num frames 2900...
+[2023-03-11 11:50:06,085][00127] Num frames 3000...
+[2023-03-11 11:50:06,194][00127] Num frames 3100...
+[2023-03-11 11:50:06,311][00127] Num frames 3200...
+[2023-03-11 11:50:06,423][00127] Num frames 3300...
+[2023-03-11 11:50:06,531][00127] Num frames 3400...
+[2023-03-11 11:50:06,661][00127] Num frames 3500...
+[2023-03-11 11:50:06,776][00127] Num frames 3600...
+[2023-03-11 11:50:06,828][00127] Avg episode rewards: #0: 27.333, true rewards: #0: 12.000
+[2023-03-11 11:50:06,830][00127] Avg episode reward: 27.333, avg true_objective: 12.000
+[2023-03-11 11:50:06,941][00127] Num frames 3700...
+[2023-03-11 11:50:07,058][00127] Num frames 3800...
+[2023-03-11 11:50:07,163][00127] Num frames 3900...
+[2023-03-11 11:50:07,270][00127] Num frames 4000...
+[2023-03-11 11:50:07,381][00127] Avg episode rewards: #0: 21.870, true rewards: #0: 10.120
+[2023-03-11 11:50:07,383][00127] Avg episode reward: 21.870, avg true_objective: 10.120
+[2023-03-11 11:50:07,452][00127] Num frames 4100...
+[2023-03-11 11:50:07,565][00127] Num frames 4200...
+[2023-03-11 11:50:07,679][00127] Num frames 4300...
+[2023-03-11 11:50:07,790][00127] Num frames 4400...
+[2023-03-11 11:50:07,908][00127] Num frames 4500...
+[2023-03-11 11:50:08,029][00127] Num frames 4600...
+[2023-03-11 11:50:08,141][00127] Num frames 4700...
+[2023-03-11 11:50:08,257][00127] Num frames 4800...
+[2023-03-11 11:50:08,378][00127] Num frames 4900...
+[2023-03-11 11:50:08,483][00127] Avg episode rewards: #0: 21.488, true rewards: #0: 9.888
+[2023-03-11 11:50:08,484][00127] Avg episode reward: 21.488, avg true_objective: 9.888
+[2023-03-11 11:50:08,553][00127] Num frames 5000...
+[2023-03-11 11:50:08,665][00127] Num frames 5100...
+[2023-03-11 11:50:08,776][00127] Num frames 5200...
+[2023-03-11 11:50:08,905][00127] Num frames 5300...
+[2023-03-11 11:50:09,033][00127] Num frames 5400...
+[2023-03-11 11:50:09,144][00127] Num frames 5500...
+[2023-03-11 11:50:09,255][00127] Num frames 5600...
+[2023-03-11 11:50:09,398][00127] Avg episode rewards: #0: 20.462, true rewards: #0: 9.462
+[2023-03-11 11:50:09,400][00127] Avg episode reward: 20.462, avg true_objective: 9.462
+[2023-03-11 11:50:09,440][00127] Num frames 5700...
+[2023-03-11 11:50:09,550][00127] Num frames 5800...
+[2023-03-11 11:50:09,662][00127] Num frames 5900...
+[2023-03-11 11:50:09,791][00127] Num frames 6000...
+[2023-03-11 11:50:09,950][00127] Num frames 6100...
+[2023-03-11 11:50:10,097][00127] Avg episode rewards: #0: 18.653, true rewards: #0: 8.796
+[2023-03-11 11:50:10,099][00127] Avg episode reward: 18.653, avg true_objective: 8.796
+[2023-03-11 11:50:10,170][00127] Num frames 6200...
+[2023-03-11 11:50:10,324][00127] Num frames 6300...
+[2023-03-11 11:50:10,479][00127] Num frames 6400...
+[2023-03-11 11:50:10,633][00127] Num frames 6500...
+[2023-03-11 11:50:10,710][00127] Avg episode rewards: #0: 16.886, true rewards: #0: 8.136
+[2023-03-11 11:50:10,714][00127] Avg episode reward: 16.886, avg true_objective: 8.136
+[2023-03-11 11:50:10,861][00127] Num frames 6600...
+[2023-03-11 11:50:11,011][00127] Num frames 6700...
+[2023-03-11 11:50:11,161][00127] Num frames 6800...
+[2023-03-11 11:50:11,320][00127] Num frames 6900...
+[2023-03-11 11:50:11,473][00127] Num frames 7000...
+[2023-03-11 11:50:11,634][00127] Num frames 7100...
+[2023-03-11 11:50:11,794][00127] Num frames 7200...
+[2023-03-11 11:50:11,969][00127] Num frames 7300...
+[2023-03-11 11:50:12,154][00127] Num frames 7400...
+[2023-03-11 11:50:12,328][00127] Num frames 7500...
+[2023-03-11 11:50:12,499][00127] Num frames 7600...
+[2023-03-11 11:50:12,663][00127] Num frames 7700...
+[2023-03-11 11:50:12,823][00127] Num frames 7800...
+[2023-03-11 11:50:12,980][00127] Num frames 7900...
+[2023-03-11 11:50:13,146][00127] Num frames 8000...
+[2023-03-11 11:50:13,323][00127] Avg episode rewards: #0: 19.860, true rewards: #0: 8.971
+[2023-03-11 11:50:13,325][00127] Avg episode reward: 19.860, avg true_objective: 8.971
+[2023-03-11 11:50:13,358][00127] Num frames 8100...
+[2023-03-11 11:50:13,470][00127] Num frames 8200...
+[2023-03-11 11:50:13,581][00127] Num frames 8300...
+[2023-03-11 11:50:13,695][00127] Num frames 8400...
+[2023-03-11 11:50:13,823][00127] Num frames 8500...
+[2023-03-11 11:50:13,944][00127] Num frames 8600...
+[2023-03-11 11:50:14,054][00127] Num frames 8700...
+[2023-03-11 11:50:14,173][00127] Num frames 8800...
+[2023-03-11 11:50:14,292][00127] Num frames 8900...
+[2023-03-11 11:50:14,404][00127] Num frames 9000...
+[2023-03-11 11:50:14,517][00127] Num frames 9100...
+[2023-03-11 11:50:14,630][00127] Num frames 9200...
+[2023-03-11 11:50:14,756][00127] Num frames 9300...
+[2023-03-11 11:50:14,865][00127] Num frames 9400...
+[2023-03-11 11:50:15,017][00127] Avg episode rewards: #0: 21.182, true rewards: #0: 9.482
+[2023-03-11 11:50:15,018][00127] Avg episode reward: 21.182, avg true_objective: 9.482
+[2023-03-11 11:51:10,872][00127] Replay video saved to /content/train_dir/default_experiment/replay.mp4!