DavidPL1 commited on
Commit
3a5597c
1 Parent(s): 35cb56f

Upload folder using huggingface_hub

Browse files
.summary/0/events.out.tfevents.1716845044.wallenstein ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:675b848754089419efcb77f6bee70a13962780b31510a6ce78dc2250f304dc8f
3
+ size 108302
README.md CHANGED
@@ -15,7 +15,7 @@ model-index:
15
  type: doom_health_gathering_supreme
16
  metrics:
17
  - type: mean_reward
18
- value: 12.63 +/- 5.37
19
  name: mean_reward
20
  verified: false
21
  ---
 
15
  type: doom_health_gathering_supreme
16
  metrics:
17
  - type: mean_reward
18
+ value: 11.14 +/- 4.80
19
  name: mean_reward
20
  verified: false
21
  ---
checkpoint_p0/best_000001779_7286784_reward_25.925.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aff07400a3c664f9d50e2c4fcb4e0e73f8ec8fa0ee27ceaf547d8678f1b78d1d
3
+ size 34929243
checkpoint_p0/checkpoint_000001779_7286784.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51623b71960451c1fc678c186a6cf68819cbbb290018bd5a4058234b5a3a0026
3
+ size 34929669
checkpoint_p0/checkpoint_000001955_8007680.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ed26c604f0d18fb4176e513a7cb452d06bd0df83e5791f1c53f6b9203f6dddb
3
+ size 34929669
config.json CHANGED
@@ -16,7 +16,7 @@
16
  "policy_workers_per_policy": 1,
17
  "max_policy_lag": 1000,
18
  "num_workers": 8,
19
- "num_envs_per_worker": 4,
20
  "batch_size": 1024,
21
  "num_batches_per_epoch": 1,
22
  "num_epochs": 1,
@@ -65,7 +65,7 @@
65
  "summaries_use_frameskip": true,
66
  "heartbeat_interval": 20,
67
  "heartbeat_reporting_interval": 600,
68
- "train_for_env_steps": 4000000,
69
  "train_for_seconds": 10000000000,
70
  "save_every_sec": 120,
71
  "keep_checkpoints": 2,
 
16
  "policy_workers_per_policy": 1,
17
  "max_policy_lag": 1000,
18
  "num_workers": 8,
19
+ "num_envs_per_worker": 8,
20
  "batch_size": 1024,
21
  "num_batches_per_epoch": 1,
22
  "num_epochs": 1,
 
65
  "summaries_use_frameskip": true,
66
  "heartbeat_interval": 20,
67
  "heartbeat_reporting_interval": 600,
68
+ "train_for_env_steps": 8000000,
69
  "train_for_seconds": 10000000000,
70
  "save_every_sec": 120,
71
  "keep_checkpoints": 2,
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbf2fc33560e2134cac2811ddcf485e499f699112159fdc74a322afb1f7cd002
3
- size 24006025
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a743a18b22b4d2c0827a27a35b96ef5dc116ec03497abba49cacaf2f3048315a
3
+ size 21230418
sf_log.txt CHANGED
@@ -781,3 +781,746 @@ main_loop: 204.3645
781
  [2024-05-27 23:20:02,394][1934158] Avg episode rewards: #0: 30.430, true rewards: #0: 12.630
782
  [2024-05-27 23:20:02,394][1934158] Avg episode reward: 30.430, avg true_objective: 12.630
783
  [2024-05-27 23:20:17,697][1934158] Replay video saved to /media/fast/code/learning/train_dir/default_experiment/replay.mp4!
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
781
  [2024-05-27 23:20:02,394][1934158] Avg episode rewards: #0: 30.430, true rewards: #0: 12.630
782
  [2024-05-27 23:20:02,394][1934158] Avg episode reward: 30.430, avg true_objective: 12.630
783
  [2024-05-27 23:20:17,697][1934158] Replay video saved to /media/fast/code/learning/train_dir/default_experiment/replay.mp4!
784
+ [2024-05-27 23:20:23,037][1934158] The model has been pushed to https://huggingface.co/DavidPL1/rl_course_vizdoom_health_gathering_supreme
785
+ [2024-05-27 23:24:06,850][1944769] Saving configuration to /media/fast/code/learning/train_dir/default_experiment/config.json...
786
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 0 uses device cpu
787
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 1 uses device cpu
788
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 2 uses device cpu
789
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 3 uses device cpu
790
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 4 uses device cpu
791
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 5 uses device cpu
792
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 6 uses device cpu
793
+ [2024-05-27 23:24:06,851][1944769] Rollout worker 7 uses device cpu
794
+ [2024-05-27 23:24:06,885][1944769] Using GPUs [0] for process 0 (actually maps to GPUs [0])
795
+ [2024-05-27 23:24:06,885][1944769] InferenceWorker_p0-w0: min num requests: 2
796
+ [2024-05-27 23:24:06,898][1944769] Starting all processes...
797
+ [2024-05-27 23:24:06,898][1944769] Starting process learner_proc0
798
+ [2024-05-27 23:24:08,668][1944769] Starting all processes...
799
+ [2024-05-27 23:24:08,671][1944769] Starting process inference_proc0-0
800
+ [2024-05-27 23:24:08,671][1944769] Starting process rollout_proc0
801
+ [2024-05-27 23:24:08,672][1944769] Starting process rollout_proc4
802
+ [2024-05-27 23:24:08,672][1944894] Using GPUs [0] for process 0 (actually maps to GPUs [0])
803
+ [2024-05-27 23:24:08,672][1944894] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
804
+ [2024-05-27 23:24:08,671][1944769] Starting process rollout_proc2
805
+ [2024-05-27 23:24:08,672][1944769] Starting process rollout_proc3
806
+ [2024-05-27 23:24:08,671][1944769] Starting process rollout_proc1
807
+ [2024-05-27 23:24:08,672][1944769] Starting process rollout_proc5
808
+ [2024-05-27 23:24:08,672][1944769] Starting process rollout_proc6
809
+ [2024-05-27 23:24:08,682][1944894] Num visible devices: 1
810
+ [2024-05-27 23:24:08,672][1944769] Starting process rollout_proc7
811
+ [2024-05-27 23:24:08,707][1944894] Starting seed is not provided
812
+ [2024-05-27 23:24:08,707][1944894] Using GPUs [0] for process 0 (actually maps to GPUs [0])
813
+ [2024-05-27 23:24:08,707][1944894] Initializing actor-critic model on device cuda:0
814
+ [2024-05-27 23:24:08,708][1944894] RunningMeanStd input shape: (3, 72, 128)
815
+ [2024-05-27 23:24:08,708][1944894] RunningMeanStd input shape: (1,)
816
+ [2024-05-27 23:24:08,721][1944894] ConvEncoder: input_channels=3
817
+ [2024-05-27 23:24:08,826][1944894] Conv encoder output size: 512
818
+ [2024-05-27 23:24:08,827][1944894] Policy head output size: 512
819
+ [2024-05-27 23:24:08,840][1944894] Created Actor Critic model with architecture:
820
+ [2024-05-27 23:24:08,840][1944894] ActorCriticSharedWeights(
821
+ (obs_normalizer): ObservationNormalizer(
822
+ (running_mean_std): RunningMeanStdDictInPlace(
823
+ (running_mean_std): ModuleDict(
824
+ (obs): RunningMeanStdInPlace()
825
+ )
826
+ )
827
+ )
828
+ (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
829
+ (encoder): VizdoomEncoder(
830
+ (basic_encoder): ConvEncoder(
831
+ (enc): RecursiveScriptModule(
832
+ original_name=ConvEncoderImpl
833
+ (conv_head): RecursiveScriptModule(
834
+ original_name=Sequential
835
+ (0): RecursiveScriptModule(original_name=Conv2d)
836
+ (1): RecursiveScriptModule(original_name=ELU)
837
+ (2): RecursiveScriptModule(original_name=Conv2d)
838
+ (3): RecursiveScriptModule(original_name=ELU)
839
+ (4): RecursiveScriptModule(original_name=Conv2d)
840
+ (5): RecursiveScriptModule(original_name=ELU)
841
+ )
842
+ (mlp_layers): RecursiveScriptModule(
843
+ original_name=Sequential
844
+ (0): RecursiveScriptModule(original_name=Linear)
845
+ (1): RecursiveScriptModule(original_name=ELU)
846
+ )
847
+ )
848
+ )
849
+ )
850
+ (core): ModelCoreRNN(
851
+ (core): GRU(512, 512)
852
+ )
853
+ (decoder): MlpDecoder(
854
+ (mlp): Identity()
855
+ )
856
+ (critic_linear): Linear(in_features=512, out_features=1, bias=True)
857
+ (action_parameterization): ActionParameterizationDefault(
858
+ (distribution_linear): Linear(in_features=512, out_features=5, bias=True)
859
+ )
860
+ )
861
+ [2024-05-27 23:24:08,935][1944894] Using optimizer <class 'torch.optim.adam.Adam'>
862
+ [2024-05-27 23:24:09,730][1944894] Loading state from checkpoint /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
863
+ [2024-05-27 23:24:09,767][1944894] Loading model from checkpoint
864
+ [2024-05-27 23:24:09,769][1944894] Loaded experiment state at self.train_step=978, self.env_steps=4005888
865
+ [2024-05-27 23:24:09,769][1944894] Initialized policy 0 weights for model version 978
866
+ [2024-05-27 23:24:09,771][1944894] LearnerWorker_p0 finished initialization!
867
+ [2024-05-27 23:24:09,771][1944894] Using GPUs [0] for process 0 (actually maps to GPUs [0])
868
+ [2024-05-27 23:24:10,329][1944942] Worker 3 uses CPU cores [6, 7]
869
+ [2024-05-27 23:24:10,513][1944941] Worker 2 uses CPU cores [4, 5]
870
+ [2024-05-27 23:24:10,610][1944938] Using GPUs [0] for process 0 (actually maps to GPUs [0])
871
+ [2024-05-27 23:24:10,610][1944938] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
872
+ [2024-05-27 23:24:10,620][1944938] Num visible devices: 1
873
+ [2024-05-27 23:24:10,718][1944940] Worker 4 uses CPU cores [8, 9]
874
+ [2024-05-27 23:24:10,777][1944959] Worker 5 uses CPU cores [10, 11]
875
+ [2024-05-27 23:24:10,791][1944960] Worker 6 uses CPU cores [12, 13]
876
+ [2024-05-27 23:24:10,810][1944939] Worker 0 uses CPU cores [0, 1]
877
+ [2024-05-27 23:24:10,825][1944769] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 4005888. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
878
+ [2024-05-27 23:24:10,826][1944943] Worker 1 uses CPU cores [2, 3]
879
+ [2024-05-27 23:24:10,829][1944961] Worker 7 uses CPU cores [14, 15]
880
+ [2024-05-27 23:24:10,925][1944938] RunningMeanStd input shape: (3, 72, 128)
881
+ [2024-05-27 23:24:10,926][1944938] RunningMeanStd input shape: (1,)
882
+ [2024-05-27 23:24:10,932][1944938] ConvEncoder: input_channels=3
883
+ [2024-05-27 23:24:10,989][1944938] Conv encoder output size: 512
884
+ [2024-05-27 23:24:10,989][1944938] Policy head output size: 512
885
+ [2024-05-27 23:24:11,020][1944769] Inference worker 0-0 is ready!
886
+ [2024-05-27 23:24:11,020][1944769] All inference workers are ready! Signal rollout workers to start!
887
+ [2024-05-27 23:24:11,031][1944941] Doom resolution: 160x120, resize resolution: (128, 72)
888
+ [2024-05-27 23:24:11,033][1944960] Doom resolution: 160x120, resize resolution: (128, 72)
889
+ [2024-05-27 23:24:11,035][1944959] Doom resolution: 160x120, resize resolution: (128, 72)
890
+ [2024-05-27 23:24:11,036][1944940] Doom resolution: 160x120, resize resolution: (128, 72)
891
+ [2024-05-27 23:24:11,036][1944943] Doom resolution: 160x120, resize resolution: (128, 72)
892
+ [2024-05-27 23:24:11,036][1944939] Doom resolution: 160x120, resize resolution: (128, 72)
893
+ [2024-05-27 23:24:11,036][1944961] Doom resolution: 160x120, resize resolution: (128, 72)
894
+ [2024-05-27 23:24:11,036][1944942] Doom resolution: 160x120, resize resolution: (128, 72)
895
+ [2024-05-27 23:24:11,635][1944940] Decorrelating experience for 0 frames...
896
+ [2024-05-27 23:24:11,716][1944959] Decorrelating experience for 0 frames...
897
+ [2024-05-27 23:24:11,795][1944943] Decorrelating experience for 0 frames...
898
+ [2024-05-27 23:24:11,796][1944941] Decorrelating experience for 0 frames...
899
+ [2024-05-27 23:24:11,798][1944960] Decorrelating experience for 0 frames...
900
+ [2024-05-27 23:24:12,598][1944940] Decorrelating experience for 32 frames...
901
+ [2024-05-27 23:24:12,600][1944959] Decorrelating experience for 32 frames...
902
+ [2024-05-27 23:24:12,679][1944942] Decorrelating experience for 0 frames...
903
+ [2024-05-27 23:24:12,679][1944961] Decorrelating experience for 0 frames...
904
+ [2024-05-27 23:24:12,681][1944960] Decorrelating experience for 32 frames...
905
+ [2024-05-27 23:24:12,997][1944941] Decorrelating experience for 32 frames...
906
+ [2024-05-27 23:24:13,512][1944940] Decorrelating experience for 64 frames...
907
+ [2024-05-27 23:24:13,592][1944942] Decorrelating experience for 32 frames...
908
+ [2024-05-27 23:24:13,594][1944943] Decorrelating experience for 32 frames...
909
+ [2024-05-27 23:24:13,594][1944959] Decorrelating experience for 64 frames...
910
+ [2024-05-27 23:24:13,997][1944939] Decorrelating experience for 0 frames...
911
+ [2024-05-27 23:24:14,001][1944960] Decorrelating experience for 64 frames...
912
+ [2024-05-27 23:24:14,080][1944941] Decorrelating experience for 64 frames...
913
+ [2024-05-27 23:24:14,331][1944769] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 4005888. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
914
+ [2024-05-27 23:24:14,411][1944942] Decorrelating experience for 64 frames...
915
+ [2024-05-27 23:24:14,490][1944959] Decorrelating experience for 96 frames...
916
+ [2024-05-27 23:24:14,729][1944940] Decorrelating experience for 96 frames...
917
+ [2024-05-27 23:24:14,811][1944939] Decorrelating experience for 32 frames...
918
+ [2024-05-27 23:24:14,889][1944943] Decorrelating experience for 64 frames...
919
+ [2024-05-27 23:24:15,357][1944942] Decorrelating experience for 96 frames...
920
+ [2024-05-27 23:24:15,358][1944960] Decorrelating experience for 96 frames...
921
+ [2024-05-27 23:24:15,359][1944959] Decorrelating experience for 128 frames...
922
+ [2024-05-27 23:24:15,681][1944961] Decorrelating experience for 32 frames...
923
+ [2024-05-27 23:24:15,765][1944940] Decorrelating experience for 128 frames...
924
+ [2024-05-27 23:24:15,843][1944943] Decorrelating experience for 96 frames...
925
+ [2024-05-27 23:24:16,252][1944939] Decorrelating experience for 64 frames...
926
+ [2024-05-27 23:24:16,419][1944959] Decorrelating experience for 160 frames...
927
+ [2024-05-27 23:24:16,419][1944960] Decorrelating experience for 128 frames...
928
+ [2024-05-27 23:24:16,674][1944942] Decorrelating experience for 128 frames...
929
+ [2024-05-27 23:24:16,759][1944941] Decorrelating experience for 96 frames...
930
+ [2024-05-27 23:24:16,760][1944961] Decorrelating experience for 64 frames...
931
+ [2024-05-27 23:24:17,276][1944939] Decorrelating experience for 96 frames...
932
+ [2024-05-27 23:24:17,359][1944940] Decorrelating experience for 160 frames...
933
+ [2024-05-27 23:24:17,592][1944960] Decorrelating experience for 160 frames...
934
+ [2024-05-27 23:24:17,676][1944959] Decorrelating experience for 192 frames...
935
+ [2024-05-27 23:24:17,764][1944961] Decorrelating experience for 96 frames...
936
+ [2024-05-27 23:24:17,764][1944943] Decorrelating experience for 128 frames...
937
+ [2024-05-27 23:24:18,096][1944939] Decorrelating experience for 128 frames...
938
+ [2024-05-27 23:24:18,410][1944940] Decorrelating experience for 192 frames...
939
+ [2024-05-27 23:24:18,572][1944960] Decorrelating experience for 192 frames...
940
+ [2024-05-27 23:24:18,648][1944959] Decorrelating experience for 224 frames...
941
+ [2024-05-27 23:24:18,649][1944941] Decorrelating experience for 128 frames...
942
+ [2024-05-27 23:24:18,649][1944961] Decorrelating experience for 128 frames...
943
+ [2024-05-27 23:24:19,132][1944942] Decorrelating experience for 160 frames...
944
+ [2024-05-27 23:24:19,288][1944939] Decorrelating experience for 160 frames...
945
+ [2024-05-27 23:24:19,291][1944960] Decorrelating experience for 224 frames...
946
+ [2024-05-27 23:24:19,331][1944769] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 4005888. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
947
+ [2024-05-27 23:24:19,332][1944769] Avg episode reward: [(0, '0.640')]
948
+ [2024-05-27 23:24:19,460][1944940] Decorrelating experience for 224 frames...
949
+ [2024-05-27 23:24:19,795][1944943] Decorrelating experience for 160 frames...
950
+ [2024-05-27 23:24:19,872][1944961] Decorrelating experience for 160 frames...
951
+ [2024-05-27 23:24:20,033][1944942] Decorrelating experience for 192 frames...
952
+ [2024-05-27 23:24:20,147][1944894] Signal inference workers to stop experience collection...
953
+ [2024-05-27 23:24:20,150][1944938] InferenceWorker_p0-w0: stopping experience collection
954
+ [2024-05-27 23:24:20,353][1944943] Decorrelating experience for 192 frames...
955
+ [2024-05-27 23:24:20,357][1944939] Decorrelating experience for 192 frames...
956
+ [2024-05-27 23:24:20,437][1944961] Decorrelating experience for 192 frames...
957
+ [2024-05-27 23:24:20,795][1944942] Decorrelating experience for 224 frames...
958
+ [2024-05-27 23:24:20,961][1944894] Signal inference workers to resume experience collection...
959
+ [2024-05-27 23:24:20,962][1944938] InferenceWorker_p0-w0: resuming experience collection
960
+ [2024-05-27 23:24:20,962][1944939] Decorrelating experience for 224 frames...
961
+ [2024-05-27 23:24:21,051][1944961] Decorrelating experience for 224 frames...
962
+ [2024-05-27 23:24:21,251][1944941] Decorrelating experience for 160 frames...
963
+ [2024-05-27 23:24:21,685][1944943] Decorrelating experience for 224 frames...
964
+ [2024-05-27 23:24:21,768][1944941] Decorrelating experience for 192 frames...
965
+ [2024-05-27 23:24:22,199][1944941] Decorrelating experience for 224 frames...
966
+ [2024-05-27 23:24:22,633][1944938] Updated weights for policy 0, policy_version 988 (0.0142)
967
+ [2024-05-27 23:24:23,973][1944938] Updated weights for policy 0, policy_version 998 (0.0009)
968
+ [2024-05-27 23:24:24,332][1944769] Fps is (10 sec: 9011.1, 60 sec: 6671.7, 300 sec: 6671.7). Total num frames: 4096000. Throughput: 0: 1034.8. Samples: 13976. Policy #0 lag: (min: 0.0, avg: 1.3, max: 4.0)
969
+ [2024-05-27 23:24:24,332][1944769] Avg episode reward: [(0, '14.106')]
970
+ [2024-05-27 23:24:25,343][1944938] Updated weights for policy 0, policy_version 1008 (0.0008)
971
+ [2024-05-27 23:24:26,541][1944938] Updated weights for policy 0, policy_version 1018 (0.0008)
972
+ [2024-05-27 23:24:26,882][1944769] Heartbeat connected on Batcher_0
973
+ [2024-05-27 23:24:26,883][1944769] Heartbeat connected on LearnerWorker_p0
974
+ [2024-05-27 23:24:26,889][1944769] Heartbeat connected on RolloutWorker_w0
975
+ [2024-05-27 23:24:26,890][1944769] Heartbeat connected on InferenceWorker_p0-w0
976
+ [2024-05-27 23:24:26,892][1944769] Heartbeat connected on RolloutWorker_w3
977
+ [2024-05-27 23:24:26,893][1944769] Heartbeat connected on RolloutWorker_w1
978
+ [2024-05-27 23:24:26,895][1944769] Heartbeat connected on RolloutWorker_w5
979
+ [2024-05-27 23:24:26,896][1944769] Heartbeat connected on RolloutWorker_w2
980
+ [2024-05-27 23:24:26,901][1944769] Heartbeat connected on RolloutWorker_w4
981
+ [2024-05-27 23:24:26,902][1944769] Heartbeat connected on RolloutWorker_w7
982
+ [2024-05-27 23:24:26,902][1944769] Heartbeat connected on RolloutWorker_w6
983
+ [2024-05-27 23:24:27,857][1944938] Updated weights for policy 0, policy_version 1028 (0.0009)
984
+ [2024-05-27 23:24:29,131][1944938] Updated weights for policy 0, policy_version 1038 (0.0008)
985
+ [2024-05-27 23:24:29,331][1944769] Fps is (10 sec: 24985.6, 60 sec: 13501.0, 300 sec: 13501.0). Total num frames: 4255744. Throughput: 0: 3344.3. Samples: 61892. Policy #0 lag: (min: 0.0, avg: 1.1, max: 3.0)
986
+ [2024-05-27 23:24:29,332][1944769] Avg episode reward: [(0, '19.936')]
987
+ [2024-05-27 23:24:30,405][1944938] Updated weights for policy 0, policy_version 1048 (0.0008)
988
+ [2024-05-27 23:24:31,658][1944938] Updated weights for policy 0, policy_version 1058 (0.0008)
989
+ [2024-05-27 23:24:32,904][1944938] Updated weights for policy 0, policy_version 1068 (0.0007)
990
+ [2024-05-27 23:24:34,244][1944938] Updated weights for policy 0, policy_version 1078 (0.0007)
991
+ [2024-05-27 23:24:34,331][1944769] Fps is (10 sec: 31949.2, 60 sec: 17425.0, 300 sec: 17425.0). Total num frames: 4415488. Throughput: 0: 3679.0. Samples: 86480. Policy #0 lag: (min: 0.0, avg: 1.0, max: 2.0)
992
+ [2024-05-27 23:24:34,332][1944769] Avg episode reward: [(0, '24.717')]
993
+ [2024-05-27 23:24:34,334][1944894] Saving new best policy, reward=24.717!
994
+ [2024-05-27 23:24:35,485][1944938] Updated weights for policy 0, policy_version 1088 (0.0008)
995
+ [2024-05-27 23:24:37,044][1944938] Updated weights for policy 0, policy_version 1098 (0.0014)
996
+ [2024-05-27 23:24:38,236][1944938] Updated weights for policy 0, policy_version 1108 (0.0008)
997
+ [2024-05-27 23:24:39,331][1944769] Fps is (10 sec: 31539.0, 60 sec: 19828.7, 300 sec: 19828.7). Total num frames: 4571136. Throughput: 0: 4656.3. Samples: 132736. Policy #0 lag: (min: 0.0, avg: 1.9, max: 3.0)
998
+ [2024-05-27 23:24:39,332][1944769] Avg episode reward: [(0, '24.188')]
999
+ [2024-05-27 23:24:39,508][1944938] Updated weights for policy 0, policy_version 1118 (0.0012)
1000
+ [2024-05-27 23:24:40,818][1944938] Updated weights for policy 0, policy_version 1128 (0.0009)
1001
+ [2024-05-27 23:24:42,144][1944938] Updated weights for policy 0, policy_version 1138 (0.0008)
1002
+ [2024-05-27 23:24:43,414][1944938] Updated weights for policy 0, policy_version 1148 (0.0008)
1003
+ [2024-05-27 23:24:44,332][1944769] Fps is (10 sec: 31538.7, 60 sec: 21637.2, 300 sec: 21637.2). Total num frames: 4730880. Throughput: 0: 5386.4. Samples: 180480. Policy #0 lag: (min: 0.0, avg: 1.0, max: 2.0)
1004
+ [2024-05-27 23:24:44,332][1944769] Avg episode reward: [(0, '23.123')]
1005
+ [2024-05-27 23:24:44,683][1944938] Updated weights for policy 0, policy_version 1158 (0.0010)
1006
+ [2024-05-27 23:24:45,942][1944938] Updated weights for policy 0, policy_version 1168 (0.0008)
1007
+ [2024-05-27 23:24:47,177][1944938] Updated weights for policy 0, policy_version 1178 (0.0007)
1008
+ [2024-05-27 23:24:48,496][1944938] Updated weights for policy 0, policy_version 1188 (0.0007)
1009
+ [2024-05-27 23:24:49,332][1944769] Fps is (10 sec: 31948.8, 60 sec: 22976.2, 300 sec: 22976.2). Total num frames: 4890624. Throughput: 0: 5315.0. Samples: 204664. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
1010
+ [2024-05-27 23:24:49,332][1944769] Avg episode reward: [(0, '21.083')]
1011
+ [2024-05-27 23:24:49,810][1944938] Updated weights for policy 0, policy_version 1198 (0.0008)
1012
+ [2024-05-27 23:24:51,091][1944938] Updated weights for policy 0, policy_version 1208 (0.0016)
1013
+ [2024-05-27 23:24:52,308][1944938] Updated weights for policy 0, policy_version 1218 (0.0009)
1014
+ [2024-05-27 23:24:53,558][1944938] Updated weights for policy 0, policy_version 1228 (0.0008)
1015
+ [2024-05-27 23:24:54,331][1944769] Fps is (10 sec: 32358.9, 60 sec: 24101.6, 300 sec: 24101.6). Total num frames: 5054464. Throughput: 0: 5811.0. Samples: 252816. Policy #0 lag: (min: 0.0, avg: 1.1, max: 3.0)
1016
+ [2024-05-27 23:24:54,332][1944769] Avg episode reward: [(0, '21.497')]
1017
+ [2024-05-27 23:24:54,819][1944938] Updated weights for policy 0, policy_version 1238 (0.0008)
1018
+ [2024-05-27 23:24:56,059][1944938] Updated weights for policy 0, policy_version 1248 (0.0009)
1019
+ [2024-05-27 23:24:57,290][1944938] Updated weights for policy 0, policy_version 1258 (0.0008)
1020
+ [2024-05-27 23:24:58,543][1944938] Updated weights for policy 0, policy_version 1268 (0.0008)
1021
+ [2024-05-27 23:24:59,331][1944769] Fps is (10 sec: 32768.1, 60 sec: 24994.9, 300 sec: 24994.9). Total num frames: 5218304. Throughput: 0: 6715.3. Samples: 302188. Policy #0 lag: (min: 0.0, avg: 1.2, max: 3.0)
1022
+ [2024-05-27 23:24:59,332][1944769] Avg episode reward: [(0, '23.600')]
1023
+ [2024-05-27 23:24:59,798][1944938] Updated weights for policy 0, policy_version 1278 (0.0010)
1024
+ [2024-05-27 23:25:01,050][1944938] Updated weights for policy 0, policy_version 1288 (0.0008)
1025
+ [2024-05-27 23:25:02,329][1944938] Updated weights for policy 0, policy_version 1298 (0.0008)
1026
+ [2024-05-27 23:25:03,696][1944938] Updated weights for policy 0, policy_version 1308 (0.0014)
1027
+ [2024-05-27 23:25:04,332][1944769] Fps is (10 sec: 32357.9, 60 sec: 25644.7, 300 sec: 25644.7). Total num frames: 5378048. Throughput: 0: 7255.6. Samples: 326504. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
1028
+ [2024-05-27 23:25:04,332][1944769] Avg episode reward: [(0, '22.440')]
1029
+ [2024-05-27 23:25:04,968][1944938] Updated weights for policy 0, policy_version 1318 (0.0008)
1030
+ [2024-05-27 23:25:06,188][1944938] Updated weights for policy 0, policy_version 1328 (0.0008)
1031
+ [2024-05-27 23:25:07,478][1944938] Updated weights for policy 0, policy_version 1338 (0.0008)
1032
+ [2024-05-27 23:25:08,699][1944938] Updated weights for policy 0, policy_version 1348 (0.0008)
1033
+ [2024-05-27 23:25:09,333][1944769] Fps is (10 sec: 32354.4, 60 sec: 26252.9, 300 sec: 26252.9). Total num frames: 5541888. Throughput: 0: 8013.2. Samples: 374580. Policy #0 lag: (min: 1.0, avg: 1.3, max: 3.0)
1034
+ [2024-05-27 23:25:09,333][1944769] Avg episode reward: [(0, '22.809')]
1035
+ [2024-05-27 23:25:09,969][1944938] Updated weights for policy 0, policy_version 1358 (0.0008)
1036
+ [2024-05-27 23:25:11,255][1944938] Updated weights for policy 0, policy_version 1368 (0.0018)
1037
+ [2024-05-27 23:25:12,552][1944938] Updated weights for policy 0, policy_version 1378 (0.0007)
1038
+ [2024-05-27 23:25:13,813][1944938] Updated weights for policy 0, policy_version 1388 (0.0008)
1039
+ [2024-05-27 23:25:14,331][1944769] Fps is (10 sec: 32358.9, 60 sec: 28262.4, 300 sec: 26701.9). Total num frames: 5701632. Throughput: 0: 8021.2. Samples: 422848. Policy #0 lag: (min: 0.0, avg: 1.3, max: 3.0)
1040
+ [2024-05-27 23:25:14,332][1944769] Avg episode reward: [(0, '24.407')]
1041
+ [2024-05-27 23:25:15,063][1944938] Updated weights for policy 0, policy_version 1398 (0.0008)
1042
+ [2024-05-27 23:25:16,369][1944938] Updated weights for policy 0, policy_version 1408 (0.0013)
1043
+ [2024-05-27 23:25:17,782][1944938] Updated weights for policy 0, policy_version 1418 (0.0011)
1044
+ [2024-05-27 23:25:19,073][1944938] Updated weights for policy 0, policy_version 1428 (0.0014)
1045
+ [2024-05-27 23:25:19,331][1944769] Fps is (10 sec: 31543.1, 60 sec: 30856.5, 300 sec: 27025.0). Total num frames: 5857280. Throughput: 0: 8008.2. Samples: 446848. Policy #0 lag: (min: 0.0, avg: 1.2, max: 3.0)
1046
+ [2024-05-27 23:25:19,332][1944769] Avg episode reward: [(0, '23.988')]
1047
+ [2024-05-27 23:25:20,331][1944938] Updated weights for policy 0, policy_version 1438 (0.0010)
1048
+ [2024-05-27 23:25:21,582][1944938] Updated weights for policy 0, policy_version 1448 (0.0007)
1049
+ [2024-05-27 23:25:22,872][1944938] Updated weights for policy 0, policy_version 1458 (0.0009)
1050
+ [2024-05-27 23:25:24,142][1944938] Updated weights for policy 0, policy_version 1468 (0.0013)
1051
+ [2024-05-27 23:25:24,331][1944769] Fps is (10 sec: 31539.1, 60 sec: 32017.1, 300 sec: 27360.0). Total num frames: 6017024. Throughput: 0: 8028.4. Samples: 494012. Policy #0 lag: (min: 0.0, avg: 1.1, max: 3.0)
1052
+ [2024-05-27 23:25:24,332][1944769] Avg episode reward: [(0, '24.169')]
1053
+ [2024-05-27 23:25:25,527][1944938] Updated weights for policy 0, policy_version 1478 (0.0008)
1054
+ [2024-05-27 23:25:26,732][1944938] Updated weights for policy 0, policy_version 1488 (0.0009)
1055
+ [2024-05-27 23:25:28,017][1944938] Updated weights for policy 0, policy_version 1498 (0.0008)
1056
+ [2024-05-27 23:25:29,185][1944938] Updated weights for policy 0, policy_version 1508 (0.0008)
1057
+ [2024-05-27 23:25:29,332][1944769] Fps is (10 sec: 32358.2, 60 sec: 32085.3, 300 sec: 27704.4). Total num frames: 6180864. Throughput: 0: 8043.0. Samples: 542416. Policy #0 lag: (min: 0.0, avg: 1.1, max: 2.0)
1058
+ [2024-05-27 23:25:29,332][1944769] Avg episode reward: [(0, '21.199')]
1059
+ [2024-05-27 23:25:30,420][1944938] Updated weights for policy 0, policy_version 1518 (0.0008)
1060
+ [2024-05-27 23:25:31,670][1944938] Updated weights for policy 0, policy_version 1528 (0.0012)
1061
+ [2024-05-27 23:25:32,874][1944938] Updated weights for policy 0, policy_version 1538 (0.0008)
1062
+ [2024-05-27 23:25:34,228][1944938] Updated weights for policy 0, policy_version 1548 (0.0013)
1063
+ [2024-05-27 23:25:34,332][1944769] Fps is (10 sec: 32358.2, 60 sec: 32085.3, 300 sec: 27958.5). Total num frames: 6340608. Throughput: 0: 8061.0. Samples: 567408. Policy #0 lag: (min: 0.0, avg: 1.3, max: 3.0)
1064
+ [2024-05-27 23:25:34,332][1944769] Avg episode reward: [(0, '25.027')]
1065
+ [2024-05-27 23:25:34,335][1944894] Saving new best policy, reward=25.027!
1066
+ [2024-05-27 23:25:35,469][1944938] Updated weights for policy 0, policy_version 1558 (0.0008)
1067
+ [2024-05-27 23:25:36,758][1944938] Updated weights for policy 0, policy_version 1568 (0.0008)
1068
+ [2024-05-27 23:25:38,071][1944938] Updated weights for policy 0, policy_version 1578 (0.0009)
1069
+ [2024-05-27 23:25:39,331][1944769] Fps is (10 sec: 31949.0, 60 sec: 32153.6, 300 sec: 28183.9). Total num frames: 6500352. Throughput: 0: 8060.3. Samples: 615528. Policy #0 lag: (min: 0.0, avg: 1.0, max: 3.0)
1070
+ [2024-05-27 23:25:39,332][1944769] Avg episode reward: [(0, '21.823')]
1071
+ [2024-05-27 23:25:39,412][1944938] Updated weights for policy 0, policy_version 1588 (0.0008)
1072
+ [2024-05-27 23:25:40,793][1944938] Updated weights for policy 0, policy_version 1598 (0.0014)
1073
+ [2024-05-27 23:25:42,055][1944938] Updated weights for policy 0, policy_version 1608 (0.0008)
1074
+ [2024-05-27 23:25:43,282][1944938] Updated weights for policy 0, policy_version 1618 (0.0010)
1075
+ [2024-05-27 23:25:44,331][1944769] Fps is (10 sec: 31949.1, 60 sec: 32153.7, 300 sec: 28385.3). Total num frames: 6660096. Throughput: 0: 8004.3. Samples: 662380. Policy #0 lag: (min: 0.0, avg: 1.1, max: 3.0)
1076
+ [2024-05-27 23:25:44,332][1944769] Avg episode reward: [(0, '22.736')]
1077
+ [2024-05-27 23:25:44,643][1944938] Updated weights for policy 0, policy_version 1628 (0.0014)
1078
+ [2024-05-27 23:25:46,073][1944938] Updated weights for policy 0, policy_version 1638 (0.0007)
1079
+ [2024-05-27 23:25:47,425][1944938] Updated weights for policy 0, policy_version 1648 (0.0008)
1080
+ [2024-05-27 23:25:48,626][1944938] Updated weights for policy 0, policy_version 1658 (0.0008)
1081
+ [2024-05-27 23:25:49,331][1944769] Fps is (10 sec: 31129.8, 60 sec: 32017.1, 300 sec: 28483.0). Total num frames: 6811648. Throughput: 0: 7951.6. Samples: 684324. Policy #0 lag: (min: 0.0, avg: 1.1, max: 2.0)
1082
+ [2024-05-27 23:25:49,332][1944769] Avg episode reward: [(0, '22.791')]
1083
+ [2024-05-27 23:25:49,975][1944938] Updated weights for policy 0, policy_version 1668 (0.0008)
1084
+ [2024-05-27 23:25:51,221][1944938] Updated weights for policy 0, policy_version 1678 (0.0009)
1085
+ [2024-05-27 23:25:52,458][1944938] Updated weights for policy 0, policy_version 1688 (0.0007)
1086
+ [2024-05-27 23:25:53,876][1944938] Updated weights for policy 0, policy_version 1698 (0.0008)
1087
+ [2024-05-27 23:25:54,331][1944769] Fps is (10 sec: 30720.0, 60 sec: 31880.5, 300 sec: 28610.8). Total num frames: 6967296. Throughput: 0: 7956.2. Samples: 732600. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
1088
+ [2024-05-27 23:25:54,332][1944769] Avg episode reward: [(0, '23.574')]
1089
+ [2024-05-27 23:25:55,258][1944938] Updated weights for policy 0, policy_version 1708 (0.0008)
1090
+ [2024-05-27 23:25:56,539][1944938] Updated weights for policy 0, policy_version 1718 (0.0012)
1091
+ [2024-05-27 23:25:57,767][1944938] Updated weights for policy 0, policy_version 1728 (0.0013)
1092
+ [2024-05-27 23:25:59,041][1944938] Updated weights for policy 0, policy_version 1738 (0.0009)
1093
+ [2024-05-27 23:25:59,331][1944769] Fps is (10 sec: 31539.1, 60 sec: 31812.3, 300 sec: 28764.6). Total num frames: 7127040. Throughput: 0: 7920.0. Samples: 779248. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
1094
+ [2024-05-27 23:25:59,332][1944769] Avg episode reward: [(0, '23.735')]
1095
+ [2024-05-27 23:26:00,279][1944938] Updated weights for policy 0, policy_version 1748 (0.0008)
1096
+ [2024-05-27 23:26:01,601][1944938] Updated weights for policy 0, policy_version 1758 (0.0010)
1097
+ [2024-05-27 23:26:02,916][1944938] Updated weights for policy 0, policy_version 1768 (0.0010)
1098
+ [2024-05-27 23:26:04,190][1944938] Updated weights for policy 0, policy_version 1778 (0.0009)
1099
+ [2024-05-27 23:26:04,332][1944769] Fps is (10 sec: 31948.3, 60 sec: 31812.3, 300 sec: 28904.9). Total num frames: 7286784. Throughput: 0: 7919.5. Samples: 803224. Policy #0 lag: (min: 0.0, avg: 1.0, max: 2.0)
1100
+ [2024-05-27 23:26:04,332][1944769] Avg episode reward: [(0, '25.925')]
1101
+ [2024-05-27 23:26:04,336][1944894] Saving /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000001779_7286784.pth...
1102
+ [2024-05-27 23:26:04,377][1944894] Removing /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000000539_2207744.pth
1103
+ [2024-05-27 23:26:04,385][1944894] Saving new best policy, reward=25.925!
1104
+ [2024-05-27 23:26:05,437][1944938] Updated weights for policy 0, policy_version 1788 (0.0010)
1105
+ [2024-05-27 23:26:06,807][1944938] Updated weights for policy 0, policy_version 1798 (0.0015)
1106
+ [2024-05-27 23:26:08,054][1944938] Updated weights for policy 0, policy_version 1808 (0.0016)
1107
+ [2024-05-27 23:26:09,297][1944938] Updated weights for policy 0, policy_version 1818 (0.0013)
1108
+ [2024-05-27 23:26:09,332][1944769] Fps is (10 sec: 31948.4, 60 sec: 31744.6, 300 sec: 29033.3). Total num frames: 7446528. Throughput: 0: 7928.8. Samples: 850808. Policy #0 lag: (min: 0.0, avg: 1.3, max: 3.0)
1109
+ [2024-05-27 23:26:09,332][1944769] Avg episode reward: [(0, '25.408')]
1110
+ [2024-05-27 23:26:10,740][1944938] Updated weights for policy 0, policy_version 1828 (0.0008)
1111
+ [2024-05-27 23:26:12,001][1944938] Updated weights for policy 0, policy_version 1838 (0.0008)
1112
+ [2024-05-27 23:26:13,281][1944938] Updated weights for policy 0, policy_version 1848 (0.0007)
1113
+ [2024-05-27 23:26:14,334][1944769] Fps is (10 sec: 31530.6, 60 sec: 31674.2, 300 sec: 29117.5). Total num frames: 7602176. Throughput: 0: 7900.8. Samples: 897972. Policy #0 lag: (min: 0.0, avg: 1.2, max: 2.0)
1114
+ [2024-05-27 23:26:14,335][1944769] Avg episode reward: [(0, '23.550')]
1115
+ [2024-05-27 23:26:14,592][1944938] Updated weights for policy 0, policy_version 1858 (0.0008)
1116
+ [2024-05-27 23:26:15,901][1944938] Updated weights for policy 0, policy_version 1868 (0.0012)
1117
+ [2024-05-27 23:26:17,177][1944938] Updated weights for policy 0, policy_version 1878 (0.0013)
1118
+ [2024-05-27 23:26:18,678][1944938] Updated weights for policy 0, policy_version 1888 (0.0014)
1119
+ [2024-05-27 23:26:19,332][1944769] Fps is (10 sec: 30720.1, 60 sec: 31607.4, 300 sec: 29164.6). Total num frames: 7753728. Throughput: 0: 7870.1. Samples: 921564. Policy #0 lag: (min: 0.0, avg: 1.2, max: 3.0)
1120
+ [2024-05-27 23:26:19,332][1944769] Avg episode reward: [(0, '24.534')]
1121
+ [2024-05-27 23:26:19,871][1944938] Updated weights for policy 0, policy_version 1898 (0.0008)
1122
+ [2024-05-27 23:26:21,197][1944938] Updated weights for policy 0, policy_version 1908 (0.0008)
1123
+ [2024-05-27 23:26:22,533][1944938] Updated weights for policy 0, policy_version 1918 (0.0009)
1124
+ [2024-05-27 23:26:23,811][1944938] Updated weights for policy 0, policy_version 1928 (0.0008)
1125
+ [2024-05-27 23:26:24,331][1944769] Fps is (10 sec: 30728.8, 60 sec: 31539.2, 300 sec: 29238.2). Total num frames: 7909376. Throughput: 0: 7826.9. Samples: 967736. Policy #0 lag: (min: 0.0, avg: 1.5, max: 4.0)
1126
+ [2024-05-27 23:26:24,332][1944769] Avg episode reward: [(0, '23.025')]
1127
+ [2024-05-27 23:26:25,114][1944938] Updated weights for policy 0, policy_version 1938 (0.0008)
1128
+ [2024-05-27 23:26:26,494][1944938] Updated weights for policy 0, policy_version 1948 (0.0015)
1129
+ [2024-05-27 23:26:27,436][1944769] Component Batcher_0 stopped!
1130
+ [2024-05-27 23:26:27,436][1944894] Stopping Batcher_0...
1131
+ [2024-05-27 23:26:27,436][1944894] Loop batcher_evt_loop terminating...
1132
+ [2024-05-27 23:26:27,437][1944894] Saving /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth...
1133
+ [2024-05-27 23:26:27,450][1944960] Stopping RolloutWorker_w6...
1134
+ [2024-05-27 23:26:27,450][1944960] Loop rollout_proc6_evt_loop terminating...
1135
+ [2024-05-27 23:26:27,450][1944769] Component RolloutWorker_w6 stopped!
1136
+ [2024-05-27 23:26:27,451][1944942] Stopping RolloutWorker_w3...
1137
+ [2024-05-27 23:26:27,451][1944769] Component RolloutWorker_w3 stopped!
1138
+ [2024-05-27 23:26:27,451][1944939] Stopping RolloutWorker_w0...
1139
+ [2024-05-27 23:26:27,452][1944942] Loop rollout_proc3_evt_loop terminating...
1140
+ [2024-05-27 23:26:27,452][1944769] Component RolloutWorker_w0 stopped!
1141
+ [2024-05-27 23:26:27,452][1944939] Loop rollout_proc0_evt_loop terminating...
1142
+ [2024-05-27 23:26:27,452][1944940] Stopping RolloutWorker_w4...
1143
+ [2024-05-27 23:26:27,452][1944769] Component RolloutWorker_w4 stopped!
1144
+ [2024-05-27 23:26:27,452][1944769] Component RolloutWorker_w7 stopped!
1145
+ [2024-05-27 23:26:27,452][1944961] Stopping RolloutWorker_w7...
1146
+ [2024-05-27 23:26:27,452][1944940] Loop rollout_proc4_evt_loop terminating...
1147
+ [2024-05-27 23:26:27,452][1944961] Loop rollout_proc7_evt_loop terminating...
1148
+ [2024-05-27 23:26:27,453][1944769] Component RolloutWorker_w5 stopped!
1149
+ [2024-05-27 23:26:27,453][1944959] Stopping RolloutWorker_w5...
1150
+ [2024-05-27 23:26:27,452][1944938] Weights refcount: 2 0
1151
+ [2024-05-27 23:26:27,453][1944959] Loop rollout_proc5_evt_loop terminating...
1152
+ [2024-05-27 23:26:27,470][1944941] Stopping RolloutWorker_w2...
1153
+ [2024-05-27 23:26:27,470][1944769] Component RolloutWorker_w2 stopped!
1154
+ [2024-05-27 23:26:27,470][1944941] Loop rollout_proc2_evt_loop terminating...
1155
+ [2024-05-27 23:26:27,482][1944894] Removing /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth
1156
+ [2024-05-27 23:26:27,489][1944894] Saving /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth...
1157
+ [2024-05-27 23:26:27,490][1944943] Stopping RolloutWorker_w1...
1158
+ [2024-05-27 23:26:27,490][1944769] Component RolloutWorker_w1 stopped!
1159
+ [2024-05-27 23:26:27,491][1944943] Loop rollout_proc1_evt_loop terminating...
1160
+ [2024-05-27 23:26:27,536][1944938] Stopping InferenceWorker_p0-w0...
1161
+ [2024-05-27 23:26:27,537][1944769] Component InferenceWorker_p0-w0 stopped!
1162
+ [2024-05-27 23:26:27,537][1944938] Loop inference_proc0-0_evt_loop terminating...
1163
+ [2024-05-27 23:26:27,566][1944894] Stopping LearnerWorker_p0...
1164
+ [2024-05-27 23:26:27,567][1944769] Component LearnerWorker_p0 stopped!
1165
+ [2024-05-27 23:26:27,567][1944894] Loop learner_proc0_evt_loop terminating...
1166
+ [2024-05-27 23:26:27,567][1944769] Waiting for process learner_proc0 to stop...
1167
+ [2024-05-27 23:26:28,282][1944769] Waiting for process inference_proc0-0 to join...
1168
+ [2024-05-27 23:26:28,282][1944769] Waiting for process rollout_proc0 to join...
1169
+ [2024-05-27 23:26:28,282][1944769] Waiting for process rollout_proc1 to join...
1170
+ [2024-05-27 23:26:28,282][1944769] Waiting for process rollout_proc2 to join...
1171
+ [2024-05-27 23:26:28,282][1944769] Waiting for process rollout_proc3 to join...
1172
+ [2024-05-27 23:26:28,283][1944769] Waiting for process rollout_proc4 to join...
1173
+ [2024-05-27 23:26:28,283][1944769] Waiting for process rollout_proc5 to join...
1174
+ [2024-05-27 23:26:28,283][1944769] Waiting for process rollout_proc6 to join...
1175
+ [2024-05-27 23:26:28,283][1944769] Waiting for process rollout_proc7 to join...
1176
+ [2024-05-27 23:26:28,283][1944769] Batcher 0 profile tree view:
1177
+ batching: 15.3245, releasing_batches: 0.0233
1178
+ [2024-05-27 23:26:28,283][1944769] InferenceWorker_p0-w0 profile tree view:
1179
+ wait_policy: 0.0000
1180
+ wait_policy_total: 9.0043
1181
+ update_model: 2.2328
1182
+ weight_update: 0.0020
1183
+ one_step: 0.0026
1184
+ handle_policy_step: 119.2006
1185
+ deserialize: 5.5536, stack: 0.5907, obs_to_device_normalize: 28.7723, forward: 62.5003, send_messages: 5.0913
1186
+ prepare_outputs: 12.6494
1187
+ to_cpu: 7.0284
1188
+ [2024-05-27 23:26:28,283][1944769] Learner 0 profile tree view:
1189
+ misc: 0.0041, prepare_batch: 7.3997
1190
+ train: 21.3461
1191
+ epoch_init: 0.0033, minibatch_init: 0.0037, losses_postprocess: 0.3643, kl_divergence: 0.4054, after_optimizer: 0.6822
1192
+ calculate_losses: 8.8365
1193
+ losses_init: 0.0019, forward_head: 0.5491, bptt_initial: 5.4800, tail: 0.5743, advantages_returns: 0.1749, losses: 1.0358
1194
+ bptt: 0.8461
1195
+ bptt_forward_core: 0.8056
1196
+ update: 10.6762
1197
+ clip: 0.5238
1198
+ [2024-05-27 23:26:28,283][1944769] RolloutWorker_w0 profile tree view:
1199
+ wait_for_trajectories: 0.0804, enqueue_policy_requests: 4.6237, env_step: 65.5304, overhead: 7.3248, complete_rollouts: 0.1038
1200
+ save_policy_outputs: 5.4512
1201
+ split_output_tensors: 2.5760
1202
+ [2024-05-27 23:26:28,283][1944769] RolloutWorker_w7 profile tree view:
1203
+ wait_for_trajectories: 0.0774, enqueue_policy_requests: 4.4785, env_step: 68.5710, overhead: 7.0470, complete_rollouts: 0.1004
1204
+ save_policy_outputs: 5.3293
1205
+ split_output_tensors: 2.4943
1206
+ [2024-05-27 23:26:28,283][1944769] Loop Runner_EvtLoop terminating...
1207
+ [2024-05-27 23:26:28,283][1944769] Runner profile tree view:
1208
+ main_loop: 141.3852
1209
+ [2024-05-27 23:26:28,284][1944769] Collected {0: 8007680}, FPS: 28304.2
1210
+ [2024-05-27 23:26:28,493][1944769] Loading existing experiment configuration from /media/fast/code/learning/train_dir/default_experiment/config.json
1211
+ [2024-05-27 23:26:28,493][1944769] Overriding arg 'num_workers' with value 1 passed from command line
1212
+ [2024-05-27 23:26:28,493][1944769] Adding new argument 'no_render'=True that is not in the saved config file!
1213
+ [2024-05-27 23:26:28,493][1944769] Adding new argument 'save_video'=True that is not in the saved config file!
1214
+ [2024-05-27 23:26:28,493][1944769] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
1215
+ [2024-05-27 23:26:28,493][1944769] Adding new argument 'video_name'=None that is not in the saved config file!
1216
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file!
1217
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
1218
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'push_to_hub'=False that is not in the saved config file!
1219
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'hf_repository'=None that is not in the saved config file!
1220
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'policy_index'=0 that is not in the saved config file!
1221
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
1222
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'train_script'=None that is not in the saved config file!
1223
+ [2024-05-27 23:26:28,494][1944769] Adding new argument 'enjoy_script'=None that is not in the saved config file!
1224
+ [2024-05-27 23:26:28,494][1944769] Using frameskip 1 and render_action_repeat=4 for evaluation
1225
+ [2024-05-27 23:26:28,500][1944769] Doom resolution: 160x120, resize resolution: (128, 72)
1226
+ [2024-05-27 23:26:28,501][1944769] RunningMeanStd input shape: (3, 72, 128)
1227
+ [2024-05-27 23:26:28,502][1944769] RunningMeanStd input shape: (1,)
1228
+ [2024-05-27 23:26:28,508][1944769] ConvEncoder: input_channels=3
1229
+ [2024-05-27 23:26:28,562][1944769] Conv encoder output size: 512
1230
+ [2024-05-27 23:26:28,563][1944769] Policy head output size: 512
1231
+ [2024-05-27 23:26:28,832][1944769] Loading state from checkpoint /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth...
1232
+ [2024-05-27 23:26:29,547][1944769] Num frames 100...
1233
+ [2024-05-27 23:26:29,608][1944769] Num frames 200...
1234
+ [2024-05-27 23:26:29,666][1944769] Num frames 300...
1235
+ [2024-05-27 23:26:29,736][1944769] Num frames 400...
1236
+ [2024-05-27 23:26:29,810][1944769] Num frames 500...
1237
+ [2024-05-27 23:26:29,918][1944769] Avg episode rewards: #0: 12.920, true rewards: #0: 5.920
1238
+ [2024-05-27 23:26:29,918][1944769] Avg episode reward: 12.920, avg true_objective: 5.920
1239
+ [2024-05-27 23:26:29,923][1944769] Num frames 600...
1240
+ [2024-05-27 23:26:29,980][1944769] Num frames 700...
1241
+ [2024-05-27 23:26:30,040][1944769] Num frames 800...
1242
+ [2024-05-27 23:26:30,094][1944769] Num frames 900...
1243
+ [2024-05-27 23:26:30,152][1944769] Num frames 1000...
1244
+ [2024-05-27 23:26:30,206][1944769] Num frames 1100...
1245
+ [2024-05-27 23:26:30,266][1944769] Num frames 1200...
1246
+ [2024-05-27 23:26:30,321][1944769] Num frames 1300...
1247
+ [2024-05-27 23:26:30,497][1944769] Avg episode rewards: #0: 14.300, true rewards: #0: 6.800
1248
+ [2024-05-27 23:26:30,497][1944769] Avg episode reward: 14.300, avg true_objective: 6.800
1249
+ [2024-05-27 23:26:30,522][1944769] Num frames 1400...
1250
+ [2024-05-27 23:26:30,580][1944769] Num frames 1500...
1251
+ [2024-05-27 23:26:30,637][1944769] Num frames 1600...
1252
+ [2024-05-27 23:26:30,696][1944769] Num frames 1700...
1253
+ [2024-05-27 23:26:30,760][1944769] Num frames 1800...
1254
+ [2024-05-27 23:26:30,825][1944769] Num frames 1900...
1255
+ [2024-05-27 23:26:30,884][1944769] Num frames 2000...
1256
+ [2024-05-27 23:26:30,980][1944769] Avg episode rewards: #0: 13.897, true rewards: #0: 6.897
1257
+ [2024-05-27 23:26:30,980][1944769] Avg episode reward: 13.897, avg true_objective: 6.897
1258
+ [2024-05-27 23:26:30,999][1944769] Num frames 2100...
1259
+ [2024-05-27 23:26:31,061][1944769] Num frames 2200...
1260
+ [2024-05-27 23:26:31,125][1944769] Num frames 2300...
1261
+ [2024-05-27 23:26:31,190][1944769] Num frames 2400...
1262
+ [2024-05-27 23:26:31,259][1944769] Num frames 2500...
1263
+ [2024-05-27 23:26:31,326][1944769] Num frames 2600...
1264
+ [2024-05-27 23:26:31,393][1944769] Num frames 2700...
1265
+ [2024-05-27 23:26:31,458][1944769] Num frames 2800...
1266
+ [2024-05-27 23:26:31,520][1944769] Num frames 2900...
1267
+ [2024-05-27 23:26:31,582][1944769] Num frames 3000...
1268
+ [2024-05-27 23:26:31,646][1944769] Num frames 3100...
1269
+ [2024-05-27 23:26:31,712][1944769] Num frames 3200...
1270
+ [2024-05-27 23:26:31,777][1944769] Num frames 3300...
1271
+ [2024-05-27 23:26:31,841][1944769] Num frames 3400...
1272
+ [2024-05-27 23:26:31,932][1944769] Avg episode rewards: #0: 18.405, true rewards: #0: 8.655
1273
+ [2024-05-27 23:26:31,932][1944769] Avg episode reward: 18.405, avg true_objective: 8.655
1274
+ [2024-05-27 23:26:31,956][1944769] Num frames 3500...
1275
+ [2024-05-27 23:26:32,011][1944769] Num frames 3600...
1276
+ [2024-05-27 23:26:32,073][1944769] Num frames 3700...
1277
+ [2024-05-27 23:26:32,129][1944769] Num frames 3800...
1278
+ [2024-05-27 23:26:32,191][1944769] Num frames 3900...
1279
+ [2024-05-27 23:26:32,256][1944769] Num frames 4000...
1280
+ [2024-05-27 23:26:32,319][1944769] Num frames 4100...
1281
+ [2024-05-27 23:26:32,383][1944769] Num frames 4200...
1282
+ [2024-05-27 23:26:32,445][1944769] Num frames 4300...
1283
+ [2024-05-27 23:26:32,508][1944769] Num frames 4400...
1284
+ [2024-05-27 23:26:32,572][1944769] Num frames 4500...
1285
+ [2024-05-27 23:26:32,633][1944769] Num frames 4600...
1286
+ [2024-05-27 23:26:32,692][1944769] Num frames 4700...
1287
+ [2024-05-27 23:26:32,755][1944769] Num frames 4800...
1288
+ [2024-05-27 23:26:32,821][1944769] Num frames 4900...
1289
+ [2024-05-27 23:26:32,880][1944769] Num frames 5000...
1290
+ [2024-05-27 23:26:32,942][1944769] Num frames 5100...
1291
+ [2024-05-27 23:26:33,004][1944769] Num frames 5200...
1292
+ [2024-05-27 23:26:33,091][1944769] Avg episode rewards: #0: 24.908, true rewards: #0: 10.508
1293
+ [2024-05-27 23:26:33,092][1944769] Avg episode reward: 24.908, avg true_objective: 10.508
1294
+ [2024-05-27 23:26:33,119][1944769] Num frames 5300...
1295
+ [2024-05-27 23:26:33,183][1944769] Num frames 5400...
1296
+ [2024-05-27 23:26:33,243][1944769] Num frames 5500...
1297
+ [2024-05-27 23:26:33,300][1944769] Num frames 5600...
1298
+ [2024-05-27 23:26:33,363][1944769] Num frames 5700...
1299
+ [2024-05-27 23:26:33,424][1944769] Num frames 5800...
1300
+ [2024-05-27 23:26:33,486][1944769] Num frames 5900...
1301
+ [2024-05-27 23:26:33,553][1944769] Num frames 6000...
1302
+ [2024-05-27 23:26:33,617][1944769] Num frames 6100...
1303
+ [2024-05-27 23:26:33,681][1944769] Num frames 6200...
1304
+ [2024-05-27 23:26:33,755][1944769] Num frames 6300...
1305
+ [2024-05-27 23:26:33,827][1944769] Num frames 6400...
1306
+ [2024-05-27 23:26:33,939][1944769] Avg episode rewards: #0: 25.983, true rewards: #0: 10.817
1307
+ [2024-05-27 23:26:33,939][1944769] Avg episode reward: 25.983, avg true_objective: 10.817
1308
+ [2024-05-27 23:26:33,947][1944769] Num frames 6500...
1309
+ [2024-05-27 23:26:34,086][1944769] Num frames 6600...
1310
+ [2024-05-27 23:26:34,206][1944769] Num frames 6700...
1311
+ [2024-05-27 23:26:34,274][1944769] Num frames 6800...
1312
+ [2024-05-27 23:26:34,346][1944769] Num frames 6900...
1313
+ [2024-05-27 23:26:34,502][1944769] Num frames 7000...
1314
+ [2024-05-27 23:26:34,609][1944769] Num frames 7100...
1315
+ [2024-05-27 23:26:34,698][1944769] Num frames 7200...
1316
+ [2024-05-27 23:26:34,814][1944769] Num frames 7300...
1317
+ [2024-05-27 23:26:34,891][1944769] Avg episode rewards: #0: 24.753, true rewards: #0: 10.467
1318
+ [2024-05-27 23:26:34,891][1944769] Avg episode reward: 24.753, avg true_objective: 10.467
1319
+ [2024-05-27 23:26:34,940][1944769] Num frames 7400...
1320
+ [2024-05-27 23:26:35,044][1944769] Num frames 7500...
1321
+ [2024-05-27 23:26:35,110][1944769] Num frames 7600...
1322
+ [2024-05-27 23:26:35,191][1944769] Num frames 7700...
1323
+ [2024-05-27 23:26:35,260][1944769] Num frames 7800...
1324
+ [2024-05-27 23:26:35,348][1944769] Num frames 7900...
1325
+ [2024-05-27 23:26:35,498][1944769] Avg episode rewards: #0: 22.929, true rewards: #0: 9.929
1326
+ [2024-05-27 23:26:35,499][1944769] Avg episode reward: 22.929, avg true_objective: 9.929
1327
+ [2024-05-27 23:26:35,538][1944769] Num frames 8000...
1328
+ [2024-05-27 23:26:35,630][1944769] Num frames 8100...
1329
+ [2024-05-27 23:26:35,731][1944769] Num frames 8200...
1330
+ [2024-05-27 23:26:35,882][1944769] Num frames 8300...
1331
+ [2024-05-27 23:26:35,971][1944769] Num frames 8400...
1332
+ [2024-05-27 23:26:36,041][1944769] Num frames 8500...
1333
+ [2024-05-27 23:26:36,140][1944769] Num frames 8600...
1334
+ [2024-05-27 23:26:36,295][1944769] Num frames 8700...
1335
+ [2024-05-27 23:26:36,377][1944769] Num frames 8800...
1336
+ [2024-05-27 23:26:36,446][1944769] Num frames 8900...
1337
+ [2024-05-27 23:26:36,514][1944769] Num frames 9000...
1338
+ [2024-05-27 23:26:36,602][1944769] Num frames 9100...
1339
+ [2024-05-27 23:26:36,670][1944769] Num frames 9200...
1340
+ [2024-05-27 23:26:36,737][1944769] Num frames 9300...
1341
+ [2024-05-27 23:26:36,826][1944769] Num frames 9400...
1342
+ [2024-05-27 23:26:36,926][1944769] Num frames 9500...
1343
+ [2024-05-27 23:26:37,023][1944769] Num frames 9600...
1344
+ [2024-05-27 23:26:37,092][1944769] Num frames 9700...
1345
+ [2024-05-27 23:26:37,160][1944769] Num frames 9800...
1346
+ [2024-05-27 23:26:37,295][1944769] Num frames 9900...
1347
+ [2024-05-27 23:26:37,435][1944769] Num frames 10000...
1348
+ [2024-05-27 23:26:37,516][1944769] Avg episode rewards: #0: 26.492, true rewards: #0: 11.159
1349
+ [2024-05-27 23:26:37,516][1944769] Avg episode reward: 26.492, avg true_objective: 11.159
1350
+ [2024-05-27 23:26:37,556][1944769] Num frames 10100...
1351
+ [2024-05-27 23:26:37,626][1944769] Num frames 10200...
1352
+ [2024-05-27 23:26:37,711][1944769] Num frames 10300...
1353
+ [2024-05-27 23:26:37,781][1944769] Num frames 10400...
1354
+ [2024-05-27 23:26:37,909][1944769] Num frames 10500...
1355
+ [2024-05-27 23:26:37,988][1944769] Num frames 10600...
1356
+ [2024-05-27 23:26:38,158][1944769] Num frames 10700...
1357
+ [2024-05-27 23:26:38,228][1944769] Num frames 10800...
1358
+ [2024-05-27 23:26:38,291][1944769] Num frames 10900...
1359
+ [2024-05-27 23:26:38,395][1944769] Num frames 11000...
1360
+ [2024-05-27 23:26:38,461][1944769] Num frames 11100...
1361
+ [2024-05-27 23:26:38,528][1944769] Num frames 11200...
1362
+ [2024-05-27 23:26:38,608][1944769] Num frames 11300...
1363
+ [2024-05-27 23:26:38,713][1944769] Num frames 11400...
1364
+ [2024-05-27 23:26:38,783][1944769] Num frames 11500...
1365
+ [2024-05-27 23:26:38,880][1944769] Num frames 11600...
1366
+ [2024-05-27 23:26:38,948][1944769] Num frames 11700...
1367
+ [2024-05-27 23:26:39,015][1944769] Num frames 11800...
1368
+ [2024-05-27 23:26:39,143][1944769] Num frames 11900...
1369
+ [2024-05-27 23:26:39,303][1944769] Num frames 12000...
1370
+ [2024-05-27 23:26:39,377][1944769] Num frames 12100...
1371
+ [2024-05-27 23:26:39,460][1944769] Avg episode rewards: #0: 28.943, true rewards: #0: 12.143
1372
+ [2024-05-27 23:26:39,460][1944769] Avg episode reward: 28.943, avg true_objective: 12.143
1373
+ [2024-05-27 23:26:54,252][1944769] Replay video saved to /media/fast/code/learning/train_dir/default_experiment/replay.mp4!
1374
+ [2024-05-27 23:26:54,444][1944769] Loading existing experiment configuration from /media/fast/code/learning/train_dir/default_experiment/config.json
1375
+ [2024-05-27 23:26:54,444][1944769] Overriding arg 'num_workers' with value 1 passed from command line
1376
+ [2024-05-27 23:26:54,444][1944769] Adding new argument 'no_render'=True that is not in the saved config file!
1377
+ [2024-05-27 23:26:54,444][1944769] Adding new argument 'save_video'=True that is not in the saved config file!
1378
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
1379
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'video_name'=None that is not in the saved config file!
1380
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
1381
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
1382
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'push_to_hub'=True that is not in the saved config file!
1383
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'hf_repository'='DavidPL1/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
1384
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'policy_index'=0 that is not in the saved config file!
1385
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
1386
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'train_script'=None that is not in the saved config file!
1387
+ [2024-05-27 23:26:54,445][1944769] Adding new argument 'enjoy_script'=None that is not in the saved config file!
1388
+ [2024-05-27 23:26:54,445][1944769] Using frameskip 1 and render_action_repeat=4 for evaluation
1389
+ [2024-05-27 23:26:54,450][1944769] RunningMeanStd input shape: (3, 72, 128)
1390
+ [2024-05-27 23:26:54,450][1944769] RunningMeanStd input shape: (1,)
1391
+ [2024-05-27 23:26:54,459][1944769] ConvEncoder: input_channels=3
1392
+ [2024-05-27 23:26:54,489][1944769] Conv encoder output size: 512
1393
+ [2024-05-27 23:26:54,489][1944769] Policy head output size: 512
1394
+ [2024-05-27 23:26:54,503][1944769] Loading state from checkpoint /media/fast/code/learning/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth...
1395
+ [2024-05-27 23:26:55,013][1944769] Num frames 100...
1396
+ [2024-05-27 23:26:55,114][1944769] Num frames 200...
1397
+ [2024-05-27 23:26:55,180][1944769] Num frames 300...
1398
+ [2024-05-27 23:26:55,321][1944769] Num frames 400...
1399
+ [2024-05-27 23:26:55,436][1944769] Num frames 500...
1400
+ [2024-05-27 23:26:55,524][1944769] Num frames 600...
1401
+ [2024-05-27 23:26:55,592][1944769] Num frames 700...
1402
+ [2024-05-27 23:26:55,655][1944769] Num frames 800...
1403
+ [2024-05-27 23:26:55,722][1944769] Num frames 900...
1404
+ [2024-05-27 23:26:55,787][1944769] Num frames 1000...
1405
+ [2024-05-27 23:26:55,855][1944769] Num frames 1100...
1406
+ [2024-05-27 23:26:55,999][1944769] Num frames 1200...
1407
+ [2024-05-27 23:26:56,088][1944769] Num frames 1300...
1408
+ [2024-05-27 23:26:56,195][1944769] Num frames 1400...
1409
+ [2024-05-27 23:26:56,273][1944769] Num frames 1500...
1410
+ [2024-05-27 23:26:56,339][1944769] Num frames 1600...
1411
+ [2024-05-27 23:26:56,449][1944769] Avg episode rewards: #0: 37.320, true rewards: #0: 16.320
1412
+ [2024-05-27 23:26:56,449][1944769] Avg episode reward: 37.320, avg true_objective: 16.320
1413
+ [2024-05-27 23:26:56,528][1944769] Num frames 1700...
1414
+ [2024-05-27 23:26:56,595][1944769] Num frames 1800...
1415
+ [2024-05-27 23:26:56,654][1944769] Num frames 1900...
1416
+ [2024-05-27 23:26:56,719][1944769] Num frames 2000...
1417
+ [2024-05-27 23:26:56,784][1944769] Num frames 2100...
1418
+ [2024-05-27 23:26:56,861][1944769] Num frames 2200...
1419
+ [2024-05-27 23:26:56,939][1944769] Avg episode rewards: #0: 24.700, true rewards: #0: 11.200
1420
+ [2024-05-27 23:26:56,939][1944769] Avg episode reward: 24.700, avg true_objective: 11.200
1421
+ [2024-05-27 23:26:56,981][1944769] Num frames 2300...
1422
+ [2024-05-27 23:26:57,056][1944769] Num frames 2400...
1423
+ [2024-05-27 23:26:57,182][1944769] Num frames 2500...
1424
+ [2024-05-27 23:26:57,381][1944769] Num frames 2600...
1425
+ [2024-05-27 23:26:57,449][1944769] Num frames 2700...
1426
+ [2024-05-27 23:26:57,614][1944769] Num frames 2800...
1427
+ [2024-05-27 23:26:57,734][1944769] Num frames 2900...
1428
+ [2024-05-27 23:26:57,849][1944769] Avg episode rewards: #0: 21.920, true rewards: #0: 9.920
1429
+ [2024-05-27 23:26:57,849][1944769] Avg episode reward: 21.920, avg true_objective: 9.920
1430
+ [2024-05-27 23:26:57,865][1944769] Num frames 3000...
1431
+ [2024-05-27 23:26:57,971][1944769] Num frames 3100...
1432
+ [2024-05-27 23:26:58,045][1944769] Num frames 3200...
1433
+ [2024-05-27 23:26:58,122][1944769] Num frames 3300...
1434
+ [2024-05-27 23:26:58,234][1944769] Num frames 3400...
1435
+ [2024-05-27 23:26:58,368][1944769] Num frames 3500...
1436
+ [2024-05-27 23:26:58,458][1944769] Num frames 3600...
1437
+ [2024-05-27 23:26:58,529][1944769] Num frames 3700...
1438
+ [2024-05-27 23:26:58,579][1944769] Avg episode rewards: #0: 20.500, true rewards: #0: 9.250
1439
+ [2024-05-27 23:26:58,580][1944769] Avg episode reward: 20.500, avg true_objective: 9.250
1440
+ [2024-05-27 23:26:58,647][1944769] Num frames 3800...
1441
+ [2024-05-27 23:26:58,709][1944769] Num frames 3900...
1442
+ [2024-05-27 23:26:58,773][1944769] Num frames 4000...
1443
+ [2024-05-27 23:26:58,913][1944769] Num frames 4100...
1444
+ [2024-05-27 23:26:58,993][1944769] Num frames 4200...
1445
+ [2024-05-27 23:26:59,062][1944769] Num frames 4300...
1446
+ [2024-05-27 23:26:59,129][1944769] Num frames 4400...
1447
+ [2024-05-27 23:26:59,197][1944769] Num frames 4500...
1448
+ [2024-05-27 23:26:59,309][1944769] Num frames 4600...
1449
+ [2024-05-27 23:26:59,379][1944769] Avg episode rewards: #0: 20.856, true rewards: #0: 9.256
1450
+ [2024-05-27 23:26:59,379][1944769] Avg episode reward: 20.856, avg true_objective: 9.256
1451
+ [2024-05-27 23:26:59,455][1944769] Num frames 4700...
1452
+ [2024-05-27 23:26:59,522][1944769] Num frames 4800...
1453
+ [2024-05-27 23:26:59,591][1944769] Num frames 4900...
1454
+ [2024-05-27 23:26:59,656][1944769] Num frames 5000...
1455
+ [2024-05-27 23:26:59,721][1944769] Num frames 5100...
1456
+ [2024-05-27 23:26:59,825][1944769] Num frames 5200...
1457
+ [2024-05-27 23:26:59,916][1944769] Num frames 5300...
1458
+ [2024-05-27 23:27:00,074][1944769] Num frames 5400...
1459
+ [2024-05-27 23:27:00,136][1944769] Num frames 5500...
1460
+ [2024-05-27 23:27:00,209][1944769] Num frames 5600...
1461
+ [2024-05-27 23:27:00,278][1944769] Num frames 5700...
1462
+ [2024-05-27 23:27:00,346][1944769] Num frames 5800...
1463
+ [2024-05-27 23:27:00,487][1944769] Num frames 5900...
1464
+ [2024-05-27 23:27:00,552][1944769] Num frames 6000...
1465
+ [2024-05-27 23:27:00,627][1944769] Num frames 6100...
1466
+ [2024-05-27 23:27:00,694][1944769] Num frames 6200...
1467
+ [2024-05-27 23:27:00,806][1944769] Avg episode rewards: #0: 23.820, true rewards: #0: 10.487
1468
+ [2024-05-27 23:27:00,806][1944769] Avg episode reward: 23.820, avg true_objective: 10.487
1469
+ [2024-05-27 23:27:00,812][1944769] Num frames 6300...
1470
+ [2024-05-27 23:27:00,902][1944769] Num frames 6400...
1471
+ [2024-05-27 23:27:01,011][1944769] Num frames 6500...
1472
+ [2024-05-27 23:27:01,075][1944769] Num frames 6600...
1473
+ [2024-05-27 23:27:01,198][1944769] Num frames 6700...
1474
+ [2024-05-27 23:27:01,263][1944769] Num frames 6800...
1475
+ [2024-05-27 23:27:01,353][1944769] Num frames 6900...
1476
+ [2024-05-27 23:27:01,615][1944769] Num frames 7000...
1477
+ [2024-05-27 23:27:01,684][1944769] Num frames 7100...
1478
+ [2024-05-27 23:27:01,753][1944769] Num frames 7200...
1479
+ [2024-05-27 23:27:01,840][1944769] Avg episode rewards: #0: 23.074, true rewards: #0: 10.360
1480
+ [2024-05-27 23:27:01,840][1944769] Avg episode reward: 23.074, avg true_objective: 10.360
1481
+ [2024-05-27 23:27:01,873][1944769] Num frames 7300...
1482
+ [2024-05-27 23:27:01,932][1944769] Num frames 7400...
1483
+ [2024-05-27 23:27:01,993][1944769] Num frames 7500...
1484
+ [2024-05-27 23:27:02,054][1944769] Num frames 7600...
1485
+ [2024-05-27 23:27:02,114][1944769] Num frames 7700...
1486
+ [2024-05-27 23:27:02,175][1944769] Num frames 7800...
1487
+ [2024-05-27 23:27:02,239][1944769] Num frames 7900...
1488
+ [2024-05-27 23:27:02,353][1944769] Avg episode rewards: #0: 22.110, true rewards: #0: 9.985
1489
+ [2024-05-27 23:27:02,353][1944769] Avg episode reward: 22.110, avg true_objective: 9.985
1490
+ [2024-05-27 23:27:02,426][1944769] Num frames 8000...
1491
+ [2024-05-27 23:27:02,489][1944769] Num frames 8100...
1492
+ [2024-05-27 23:27:02,544][1944769] Num frames 8200...
1493
+ [2024-05-27 23:27:02,602][1944769] Num frames 8300...
1494
+ [2024-05-27 23:27:02,663][1944769] Num frames 8400...
1495
+ [2024-05-27 23:27:02,721][1944769] Num frames 8500...
1496
+ [2024-05-27 23:27:02,784][1944769] Num frames 8600...
1497
+ [2024-05-27 23:27:02,843][1944769] Num frames 8700...
1498
+ [2024-05-27 23:27:02,904][1944769] Num frames 8800...
1499
+ [2024-05-27 23:27:02,965][1944769] Num frames 8900...
1500
+ [2024-05-27 23:27:03,028][1944769] Num frames 9000...
1501
+ [2024-05-27 23:27:03,102][1944769] Avg episode rewards: #0: 22.486, true rewards: #0: 10.041
1502
+ [2024-05-27 23:27:03,102][1944769] Avg episode reward: 22.486, avg true_objective: 10.041
1503
+ [2024-05-27 23:27:03,141][1944769] Num frames 9100...
1504
+ [2024-05-27 23:27:03,204][1944769] Num frames 9200...
1505
+ [2024-05-27 23:27:03,267][1944769] Num frames 9300...
1506
+ [2024-05-27 23:27:03,328][1944769] Num frames 9400...
1507
+ [2024-05-27 23:27:03,387][1944769] Num frames 9500...
1508
+ [2024-05-27 23:27:03,447][1944769] Num frames 9600...
1509
+ [2024-05-27 23:27:03,506][1944769] Num frames 9700...
1510
+ [2024-05-27 23:27:03,565][1944769] Num frames 9800...
1511
+ [2024-05-27 23:27:03,626][1944769] Num frames 9900...
1512
+ [2024-05-27 23:27:03,682][1944769] Num frames 10000...
1513
+ [2024-05-27 23:27:03,739][1944769] Num frames 10100...
1514
+ [2024-05-27 23:27:03,799][1944769] Num frames 10200...
1515
+ [2024-05-27 23:27:03,864][1944769] Num frames 10300...
1516
+ [2024-05-27 23:27:03,928][1944769] Num frames 10400...
1517
+ [2024-05-27 23:27:03,990][1944769] Num frames 10500...
1518
+ [2024-05-27 23:27:04,054][1944769] Num frames 10600...
1519
+ [2024-05-27 23:27:04,116][1944769] Num frames 10700...
1520
+ [2024-05-27 23:27:04,179][1944769] Num frames 10800...
1521
+ [2024-05-27 23:27:04,238][1944769] Num frames 10900...
1522
+ [2024-05-27 23:27:04,305][1944769] Num frames 11000...
1523
+ [2024-05-27 23:27:04,369][1944769] Num frames 11100...
1524
+ [2024-05-27 23:27:04,446][1944769] Avg episode rewards: #0: 25.637, true rewards: #0: 11.137
1525
+ [2024-05-27 23:27:04,446][1944769] Avg episode reward: 25.637, avg true_objective: 11.137
1526
+ [2024-05-27 23:27:18,008][1944769] Replay video saved to /media/fast/code/learning/train_dir/default_experiment/replay.mp4!