Hugging Face's logo

EntropicLettuce
/
rl_course_vizdoom_health_gathering_supreme

Reinforcement Learning
sample-factory
TensorBoard
deep-reinforcement-learning
Eval Results
Model card Files Metrics Community
EntropicLettuce commited on
Commit
59631f1
·
verified ·
1 Parent(s): 65c6280

Upload folder using huggingface_hub

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. .summary/0/events.out.tfevents.1728359045.0b81713d090a +3 -0
  3. README.md +56 -0
  4. checkpoint_p0/best_000000926_3792896_reward_22.854.pth +3 -0
  5. checkpoint_p0/checkpoint_000000527_2158592.pth +3 -0
  6. checkpoint_p0/checkpoint_000000978_4005888.pth +3 -0
  7. config.json +142 -0
  8. replay.mp4 +3 -0
  9. sf_log.txt +752 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ replay.mp4 filter=lfs diff=lfs merge=lfs -text
.summary/0/events.out.tfevents.1728359045.0b81713d090a ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:331990d96b01c6978781b1ff25187a8c5f2ef9cb5bc5e240ffaef9c37393669e
3
+ size 198418
README.md ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: sample-factory
3
+ tags:
4
+ - deep-reinforcement-learning
5
+ - reinforcement-learning
6
+ - sample-factory
7
+ model-index:
8
+ - name: APPO
9
+ results:
10
+ - task:
11
+ type: reinforcement-learning
12
+ name: reinforcement-learning
13
+ dataset:
14
+ name: doom_health_gathering_supreme
15
+ type: doom_health_gathering_supreme
16
+ metrics:
17
+ - type: mean_reward
18
+ value: 13.82 +/- 6.65
19
+ name: mean_reward
20
+ verified: false
21
+ ---
22
+
23
+ A(n) **APPO** model trained on the **doom_health_gathering_supreme** environment.
24
+
25
+ This model was trained using Sample-Factory 2.0: https://github.com/alex-petrenko/sample-factory.
26
+ Documentation for how to use Sample-Factory can be found at https://www.samplefactory.dev/
27
+
28
+
29
+ ## Downloading the model
30
+
31
+ After installing Sample-Factory, download the model with:
32
+ ```
33
+ python -m sample_factory.huggingface.load_from_hub -r EntropicLettuce/rl_course_vizdoom_health_gathering_supreme
34
+ ```
35
+
36
+
37
+ ## Using the model
38
+
39
+ To run the model after download, use the `enjoy` script corresponding to this environment:
40
+ ```
41
+ python -m <path.to.enjoy.module> --algo=APPO --env=doom_health_gathering_supreme --train_dir=./train_dir --experiment=rl_course_vizdoom_health_gathering_supreme
42
+ ```
43
+
44
+
45
+ You can also upload models to the Hugging Face Hub using the same script with the `--push_to_hub` flag.
46
+ See https://www.samplefactory.dev/10-huggingface/huggingface/ for more details
47
+
48
+ ## Training with this model
49
+
50
+ To continue training with this model, use the `train` script corresponding to this environment:
51
+ ```
52
+ python -m <path.to.train.module> --algo=APPO --env=doom_health_gathering_supreme --train_dir=./train_dir --experiment=rl_course_vizdoom_health_gathering_supreme --restart_behavior=resume --train_for_env_steps=10000000000
53
+ ```
54
+
55
+ Note, you may have to adjust `--train_for_env_steps` to a suitably high number as the experiment will resume at the number of steps it concluded at.
56
+
checkpoint_p0/best_000000926_3792896_reward_22.854.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0763b915921a3741b4a874732ed8f107eb2d99bb100f4c387d97c4a1c36373c9
3
+ size 34928614
checkpoint_p0/checkpoint_000000527_2158592.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98c765bdeb3c45ec62760c101ee3b5097d7e0cbf301660d927d2adeaa6aeb15d
3
+ size 34929028
checkpoint_p0/checkpoint_000000978_4005888.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:425d013ffdfb9bc6d0c7ed711c2b0ea9e4d8d4ae831e978ee96b93d1b33881a5
3
+ size 34929028
config.json ADDED
@@ -0,0 +1,142 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "help": false,
3
+ "algo": "APPO",
4
+ "env": "doom_health_gathering_supreme",
5
+ "experiment": "default_experiment",
6
+ "train_dir": "/content/train_dir",
7
+ "restart_behavior": "resume",
8
+ "device": "gpu",
9
+ "seed": null,
10
+ "num_policies": 1,
11
+ "async_rl": true,
12
+ "serial_mode": false,
13
+ "batched_sampling": false,
14
+ "num_batches_to_accumulate": 2,
15
+ "worker_num_splits": 2,
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,
23
+ "rollout": 32,
24
+ "recurrence": 32,
25
+ "shuffle_minibatches": false,
26
+ "gamma": 0.99,
27
+ "reward_scale": 1.0,
28
+ "reward_clip": 1000.0,
29
+ "value_bootstrap": false,
30
+ "normalize_returns": true,
31
+ "exploration_loss_coeff": 0.001,
32
+ "value_loss_coeff": 0.5,
33
+ "kl_loss_coeff": 0.0,
34
+ "exploration_loss": "symmetric_kl",
35
+ "gae_lambda": 0.95,
36
+ "ppo_clip_ratio": 0.1,
37
+ "ppo_clip_value": 0.2,
38
+ "with_vtrace": false,
39
+ "vtrace_rho": 1.0,
40
+ "vtrace_c": 1.0,
41
+ "optimizer": "adam",
42
+ "adam_eps": 1e-06,
43
+ "adam_beta1": 0.9,
44
+ "adam_beta2": 0.999,
45
+ "max_grad_norm": 4.0,
46
+ "learning_rate": 0.0001,
47
+ "lr_schedule": "constant",
48
+ "lr_schedule_kl_threshold": 0.008,
49
+ "lr_adaptive_min": 1e-06,
50
+ "lr_adaptive_max": 0.01,
51
+ "obs_subtract_mean": 0.0,
52
+ "obs_scale": 255.0,
53
+ "normalize_input": true,
54
+ "normalize_input_keys": null,
55
+ "decorrelate_experience_max_seconds": 0,
56
+ "decorrelate_envs_on_one_worker": true,
57
+ "actor_worker_gpus": [],
58
+ "set_workers_cpu_affinity": true,
59
+ "force_envs_single_thread": false,
60
+ "default_niceness": 0,
61
+ "log_to_file": true,
62
+ "experiment_summaries_interval": 10,
63
+ "flush_summaries_interval": 30,
64
+ "stats_avg": 100,
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,
72
+ "load_checkpoint_kind": "latest",
73
+ "save_milestones_sec": -1,
74
+ "save_best_every_sec": 5,
75
+ "save_best_metric": "reward",
76
+ "save_best_after": 100000,
77
+ "benchmark": false,
78
+ "encoder_mlp_layers": [
79
+ 512,
80
+ 512
81
+ ],
82
+ "encoder_conv_architecture": "convnet_simple",
83
+ "encoder_conv_mlp_layers": [
84
+ 512
85
+ ],
86
+ "use_rnn": true,
87
+ "rnn_size": 512,
88
+ "rnn_type": "gru",
89
+ "rnn_num_layers": 1,
90
+ "decoder_mlp_layers": [],
91
+ "nonlinearity": "elu",
92
+ "policy_initialization": "orthogonal",
93
+ "policy_init_gain": 1.0,
94
+ "actor_critic_share_weights": true,
95
+ "adaptive_stddev": true,
96
+ "continuous_tanh_scale": 0.0,
97
+ "initial_stddev": 1.0,
98
+ "use_env_info_cache": false,
99
+ "env_gpu_actions": false,
100
+ "env_gpu_observations": true,
101
+ "env_frameskip": 4,
102
+ "env_framestack": 1,
103
+ "pixel_format": "CHW",
104
+ "use_record_episode_statistics": false,
105
+ "with_wandb": false,
106
+ "wandb_user": null,
107
+ "wandb_project": "sample_factory",
108
+ "wandb_group": null,
109
+ "wandb_job_type": "SF",
110
+ "wandb_tags": [],
111
+ "with_pbt": false,
112
+ "pbt_mix_policies_in_one_env": true,
113
+ "pbt_period_env_steps": 5000000,
114
+ "pbt_start_mutation": 20000000,
115
+ "pbt_replace_fraction": 0.3,
116
+ "pbt_mutation_rate": 0.15,
117
+ "pbt_replace_reward_gap": 0.1,
118
+ "pbt_replace_reward_gap_absolute": 1e-06,
119
+ "pbt_optimize_gamma": false,
120
+ "pbt_target_objective": "true_objective",
121
+ "pbt_perturb_min": 1.1,
122
+ "pbt_perturb_max": 1.5,
123
+ "num_agents": -1,
124
+ "num_humans": 0,
125
+ "num_bots": -1,
126
+ "start_bot_difficulty": null,
127
+ "timelimit": null,
128
+ "res_w": 128,
129
+ "res_h": 72,
130
+ "wide_aspect_ratio": false,
131
+ "eval_env_frameskip": 1,
132
+ "fps": 35,
133
+ "command_line": "--env=doom_health_gathering_supreme --num_workers=8 --num_envs_per_worker=4 --train_for_env_steps=4000000",
134
+ "cli_args": {
135
+ "env": "doom_health_gathering_supreme",
136
+ "num_workers": 8,
137
+ "num_envs_per_worker": 4,
138
+ "train_for_env_steps": 4000000
139
+ },
140
+ "git_hash": "unknown",
141
+ "git_repo_name": "not a git repository"
142
+ }
replay.mp4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b67b75aa4fab283ab6cd0ee65d13fe372c7355c60373b68dfd7a6f1ea41cbdaf
3
+ size 26575514
sf_log.txt ADDED
@@ -0,0 +1,752 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2024-10-08 03:44:08,357][04317] Saving configuration to /content/train_dir/default_experiment/config.json...
2
+ [2024-10-08 03:44:08,359][04317] Rollout worker 0 uses device cpu
3
+ [2024-10-08 03:44:08,360][04317] Rollout worker 1 uses device cpu
4
+ [2024-10-08 03:44:08,362][04317] Rollout worker 2 uses device cpu
5
+ [2024-10-08 03:44:08,363][04317] Rollout worker 3 uses device cpu
6
+ [2024-10-08 03:44:08,365][04317] Rollout worker 4 uses device cpu
7
+ [2024-10-08 03:44:08,366][04317] Rollout worker 5 uses device cpu
8
+ [2024-10-08 03:44:08,367][04317] Rollout worker 6 uses device cpu
9
+ [2024-10-08 03:44:08,368][04317] Rollout worker 7 uses device cpu
10
+ [2024-10-08 03:44:08,423][04317] Using GPUs [0] for process 0 (actually maps to GPUs [0])
11
+ [2024-10-08 03:44:08,424][04317] InferenceWorker_p0-w0: min num requests: 2
12
+ [2024-10-08 03:44:08,458][04317] Starting all processes...
13
+ [2024-10-08 03:44:08,459][04317] Starting process learner_proc0
14
+ [2024-10-08 03:44:08,508][04317] Starting all processes...
15
+ [2024-10-08 03:44:08,513][04317] Starting process inference_proc0-0
16
+ [2024-10-08 03:44:08,514][04317] Starting process rollout_proc0
17
+ [2024-10-08 03:44:08,516][04317] Starting process rollout_proc1
18
+ [2024-10-08 03:44:08,520][04317] Starting process rollout_proc2
19
+ [2024-10-08 03:44:08,520][04317] Starting process rollout_proc3
20
+ [2024-10-08 03:44:08,521][04317] Starting process rollout_proc4
21
+ [2024-10-08 03:44:08,521][04317] Starting process rollout_proc5
22
+ [2024-10-08 03:44:08,521][04317] Starting process rollout_proc6
23
+ [2024-10-08 03:44:08,521][04317] Starting process rollout_proc7
24
+ [2024-10-08 03:44:10,457][07032] Worker 6 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
25
+ [2024-10-08 03:44:10,614][07027] Worker 0 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
26
+ [2024-10-08 03:44:10,633][07031] Worker 5 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
27
+ [2024-10-08 03:44:10,680][07008] Using GPUs [0] for process 0 (actually maps to GPUs [0])
28
+ [2024-10-08 03:44:10,681][07008] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
29
+ [2024-10-08 03:44:10,699][07008] Num visible devices: 1
30
+ [2024-10-08 03:44:10,731][07008] Starting seed is not provided
31
+ [2024-10-08 03:44:10,732][07008] Using GPUs [0] for process 0 (actually maps to GPUs [0])
32
+ [2024-10-08 03:44:10,732][07008] Initializing actor-critic model on device cuda:0
33
+ [2024-10-08 03:44:10,732][07008] RunningMeanStd input shape: (3, 72, 128)
34
+ [2024-10-08 03:44:10,734][07008] RunningMeanStd input shape: (1,)
35
+ [2024-10-08 03:44:10,754][07008] ConvEncoder: input_channels=3
36
+ [2024-10-08 03:44:10,764][07029] Worker 2 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
37
+ [2024-10-08 03:44:10,803][07021] Using GPUs [0] for process 0 (actually maps to GPUs [0])
38
+ [2024-10-08 03:44:10,803][07021] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
39
+ [2024-10-08 03:44:10,810][07022] Worker 1 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
40
+ [2024-10-08 03:44:10,819][07021] Num visible devices: 1
41
+ [2024-10-08 03:44:10,846][07030] Worker 4 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
42
+ [2024-10-08 03:44:10,909][07044] Worker 7 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
43
+ [2024-10-08 03:44:10,940][07028] Worker 3 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
44
+ [2024-10-08 03:44:10,947][07008] Conv encoder output size: 512
45
+ [2024-10-08 03:44:10,948][07008] Policy head output size: 512
46
+ [2024-10-08 03:44:10,965][07008] Created Actor Critic model with architecture:
47
+ [2024-10-08 03:44:10,965][07008] ActorCriticSharedWeights(
48
+ (obs_normalizer): ObservationNormalizer(
49
+ (running_mean_std): RunningMeanStdDictInPlace(
50
+ (running_mean_std): ModuleDict(
51
+ (obs): RunningMeanStdInPlace()
52
+ )
53
+ )
54
+ )
55
+ (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
56
+ (encoder): VizdoomEncoder(
57
+ (basic_encoder): ConvEncoder(
58
+ (enc): RecursiveScriptModule(
59
+ original_name=ConvEncoderImpl
60
+ (conv_head): RecursiveScriptModule(
61
+ original_name=Sequential
62
+ (0): RecursiveScriptModule(original_name=Conv2d)
63
+ (1): RecursiveScriptModule(original_name=ELU)
64
+ (2): RecursiveScriptModule(original_name=Conv2d)
65
+ (3): RecursiveScriptModule(original_name=ELU)
66
+ (4): RecursiveScriptModule(original_name=Conv2d)
67
+ (5): RecursiveScriptModule(original_name=ELU)
68
+ )
69
+ (mlp_layers): RecursiveScriptModule(
70
+ original_name=Sequential
71
+ (0): RecursiveScriptModule(original_name=Linear)
72
+ (1): RecursiveScriptModule(original_name=ELU)
73
+ )
74
+ )
75
+ )
76
+ )
77
+ (core): ModelCoreRNN(
78
+ (core): GRU(512, 512)
79
+ )
80
+ (decoder): MlpDecoder(
81
+ (mlp): Identity()
82
+ )
83
+ (critic_linear): Linear(in_features=512, out_features=1, bias=True)
84
+ (action_parameterization): ActionParameterizationDefault(
85
+ (distribution_linear): Linear(in_features=512, out_features=5, bias=True)
86
+ )
87
+ )
88
+ [2024-10-08 03:44:12,800][07008] Using optimizer <class 'torch.optim.adam.Adam'>
89
+ [2024-10-08 03:44:12,800][07008] No checkpoints found
90
+ [2024-10-08 03:44:12,801][07008] Did not load from checkpoint, starting from scratch!
91
+ [2024-10-08 03:44:12,801][07008] Initialized policy 0 weights for model version 0
92
+ [2024-10-08 03:44:12,803][07008] LearnerWorker_p0 finished initialization!
93
+ [2024-10-08 03:44:12,803][07008] Using GPUs [0] for process 0 (actually maps to GPUs [0])
94
+ [2024-10-08 03:44:12,878][07021] RunningMeanStd input shape: (3, 72, 128)
95
+ [2024-10-08 03:44:12,879][07021] RunningMeanStd input shape: (1,)
96
+ [2024-10-08 03:44:12,891][07021] ConvEncoder: input_channels=3
97
+ [2024-10-08 03:44:12,995][07021] Conv encoder output size: 512
98
+ [2024-10-08 03:44:12,995][07021] Policy head output size: 512
99
+ [2024-10-08 03:44:14,748][04317] Inference worker 0-0 is ready!
100
+ [2024-10-08 03:44:14,749][04317] All inference workers are ready! Signal rollout workers to start!
101
+ [2024-10-08 03:44:14,763][07031] Doom resolution: 160x120, resize resolution: (128, 72)
102
+ [2024-10-08 03:44:14,764][07032] Doom resolution: 160x120, resize resolution: (128, 72)
103
+ [2024-10-08 03:44:14,764][07030] Doom resolution: 160x120, resize resolution: (128, 72)
104
+ [2024-10-08 03:44:14,764][07028] Doom resolution: 160x120, resize resolution: (128, 72)
105
+ [2024-10-08 03:44:14,768][07029] Doom resolution: 160x120, resize resolution: (128, 72)
106
+ [2024-10-08 03:44:14,768][07027] Doom resolution: 160x120, resize resolution: (128, 72)
107
+ [2024-10-08 03:44:14,769][07022] Doom resolution: 160x120, resize resolution: (128, 72)
108
+ [2024-10-08 03:44:14,769][07044] Doom resolution: 160x120, resize resolution: (128, 72)
109
+ [2024-10-08 03:44:15,090][07029] Decorrelating experience for 0 frames...
110
+ [2024-10-08 03:44:15,090][07032] Decorrelating experience for 0 frames...
111
+ [2024-10-08 03:44:15,090][07022] Decorrelating experience for 0 frames...
112
+ [2024-10-08 03:44:15,091][07027] Decorrelating experience for 0 frames...
113
+ [2024-10-08 03:44:15,091][07031] Decorrelating experience for 0 frames...
114
+ [2024-10-08 03:44:15,091][07030] Decorrelating experience for 0 frames...
115
+ [2024-10-08 03:44:15,341][07028] Decorrelating experience for 0 frames...
116
+ [2024-10-08 03:44:15,354][07022] Decorrelating experience for 32 frames...
117
+ [2024-10-08 03:44:15,355][07029] Decorrelating experience for 32 frames...
118
+ [2024-10-08 03:44:15,357][07027] Decorrelating experience for 32 frames...
119
+ [2024-10-08 03:44:15,358][07044] Decorrelating experience for 0 frames...
120
+ [2024-10-08 03:44:15,505][04317] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 0. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
121
+ [2024-10-08 03:44:15,590][07030] Decorrelating experience for 32 frames...
122
+ [2024-10-08 03:44:15,610][07044] Decorrelating experience for 32 frames...
123
+ [2024-10-08 03:44:15,612][07028] Decorrelating experience for 32 frames...
124
+ [2024-10-08 03:44:15,643][07032] Decorrelating experience for 32 frames...
125
+ [2024-10-08 03:44:15,671][07022] Decorrelating experience for 64 frames...
126
+ [2024-10-08 03:44:15,767][07031] Decorrelating experience for 32 frames...
127
+ [2024-10-08 03:44:15,788][07027] Decorrelating experience for 64 frames...
128
+ [2024-10-08 03:44:15,883][07029] Decorrelating experience for 64 frames...
129
+ [2024-10-08 03:44:15,917][07028] Decorrelating experience for 64 frames...
130
+ [2024-10-08 03:44:15,919][07044] Decorrelating experience for 64 frames...
131
+ [2024-10-08 03:44:15,962][07022] Decorrelating experience for 96 frames...
132
+ [2024-10-08 03:44:16,082][07031] Decorrelating experience for 64 frames...
133
+ [2024-10-08 03:44:16,099][07027] Decorrelating experience for 96 frames...
134
+ [2024-10-08 03:44:16,160][07030] Decorrelating experience for 64 frames...
135
+ [2024-10-08 03:44:16,185][07029] Decorrelating experience for 96 frames...
136
+ [2024-10-08 03:44:16,377][07031] Decorrelating experience for 96 frames...
137
+ [2024-10-08 03:44:16,417][07028] Decorrelating experience for 96 frames...
138
+ [2024-10-08 03:44:16,446][07030] Decorrelating experience for 96 frames...
139
+ [2024-10-08 03:44:16,472][07044] Decorrelating experience for 96 frames...
140
+ [2024-10-08 03:44:16,727][07032] Decorrelating experience for 64 frames...
141
+ [2024-10-08 03:44:17,117][07032] Decorrelating experience for 96 frames...
142
+ [2024-10-08 03:44:17,315][07008] Signal inference workers to stop experience collection...
143
+ [2024-10-08 03:44:17,330][07021] InferenceWorker_p0-w0: stopping experience collection
144
+ [2024-10-08 03:44:18,345][07008] Signal inference workers to resume experience collection...
145
+ [2024-10-08 03:44:18,346][07021] InferenceWorker_p0-w0: resuming experience collection
146
+ [2024-10-08 03:44:20,505][04317] Fps is (10 sec: 6553.6, 60 sec: 6553.6, 300 sec: 6553.6). Total num frames: 32768. Throughput: 0: 592.8. Samples: 2964. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
147
+ [2024-10-08 03:44:20,508][04317] Avg episode reward: [(0, '3.746')]
148
+ [2024-10-08 03:44:20,783][07021] Updated weights for policy 0, policy_version 10 (0.0369)
149
+ [2024-10-08 03:44:22,786][07021] Updated weights for policy 0, policy_version 20 (0.0012)
150
+ [2024-10-08 03:44:24,901][07021] Updated weights for policy 0, policy_version 30 (0.0012)
151
+ [2024-10-08 03:44:25,505][04317] Fps is (10 sec: 13516.8, 60 sec: 13516.8, 300 sec: 13516.8). Total num frames: 135168. Throughput: 0: 2542.6. Samples: 25426. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
152
+ [2024-10-08 03:44:25,507][04317] Avg episode reward: [(0, '4.579')]
153
+ [2024-10-08 03:44:25,515][07008] Saving new best policy, reward=4.579!
154
+ [2024-10-08 03:44:26,869][07021] Updated weights for policy 0, policy_version 40 (0.0012)
155
+ [2024-10-08 03:44:28,417][04317] Heartbeat connected on Batcher_0
156
+ [2024-10-08 03:44:28,430][04317] Heartbeat connected on InferenceWorker_p0-w0
157
+ [2024-10-08 03:44:28,432][04317] Heartbeat connected on RolloutWorker_w0
158
+ [2024-10-08 03:44:28,436][04317] Heartbeat connected on RolloutWorker_w1
159
+ [2024-10-08 03:44:28,440][04317] Heartbeat connected on RolloutWorker_w2
160
+ [2024-10-08 03:44:28,443][04317] Heartbeat connected on RolloutWorker_w3
161
+ [2024-10-08 03:44:28,446][04317] Heartbeat connected on LearnerWorker_p0
162
+ [2024-10-08 03:44:28,448][04317] Heartbeat connected on RolloutWorker_w4
163
+ [2024-10-08 03:44:28,452][04317] Heartbeat connected on RolloutWorker_w5
164
+ [2024-10-08 03:44:28,454][04317] Heartbeat connected on RolloutWorker_w6
165
+ [2024-10-08 03:44:28,459][04317] Heartbeat connected on RolloutWorker_w7
166
+ [2024-10-08 03:44:28,843][07021] Updated weights for policy 0, policy_version 50 (0.0012)
167
+ [2024-10-08 03:44:30,505][04317] Fps is (10 sec: 20480.0, 60 sec: 15837.9, 300 sec: 15837.9). Total num frames: 237568. Throughput: 0: 3761.3. Samples: 56420. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
168
+ [2024-10-08 03:44:30,507][04317] Avg episode reward: [(0, '4.508')]
169
+ [2024-10-08 03:44:30,856][07021] Updated weights for policy 0, policy_version 60 (0.0011)
170
+ [2024-10-08 03:44:32,980][07021] Updated weights for policy 0, policy_version 70 (0.0012)
171
+ [2024-10-08 03:44:35,035][07021] Updated weights for policy 0, policy_version 80 (0.0011)
172
+ [2024-10-08 03:44:35,505][04317] Fps is (10 sec: 20070.1, 60 sec: 16793.5, 300 sec: 16793.5). Total num frames: 335872. Throughput: 0: 3565.4. Samples: 71308. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
173
+ [2024-10-08 03:44:35,508][04317] Avg episode reward: [(0, '4.285')]
174
+ [2024-10-08 03:44:37,035][07021] Updated weights for policy 0, policy_version 90 (0.0011)
175
+ [2024-10-08 03:44:39,028][07021] Updated weights for policy 0, policy_version 100 (0.0011)
176
+ [2024-10-08 03:44:40,505][04317] Fps is (10 sec: 20070.4, 60 sec: 17530.9, 300 sec: 17530.9). Total num frames: 438272. Throughput: 0: 4072.9. Samples: 101822. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
177
+ [2024-10-08 03:44:40,508][04317] Avg episode reward: [(0, '4.472')]
178
+ [2024-10-08 03:44:41,013][07021] Updated weights for policy 0, policy_version 110 (0.0011)
179
+ [2024-10-08 03:44:42,979][07021] Updated weights for policy 0, policy_version 120 (0.0011)
180
+ [2024-10-08 03:44:45,049][07021] Updated weights for policy 0, policy_version 130 (0.0011)
181
+ [2024-10-08 03:44:45,505][04317] Fps is (10 sec: 20480.4, 60 sec: 18022.4, 300 sec: 18022.4). Total num frames: 540672. Throughput: 0: 4413.2. Samples: 132396. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
182
+ [2024-10-08 03:44:45,507][04317] Avg episode reward: [(0, '4.779')]
183
+ [2024-10-08 03:44:45,516][07008] Saving new best policy, reward=4.779!
184
+ [2024-10-08 03:44:47,184][07021] Updated weights for policy 0, policy_version 140 (0.0012)
185
+ [2024-10-08 03:44:49,265][07021] Updated weights for policy 0, policy_version 150 (0.0011)
186
+ [2024-10-08 03:44:50,505][04317] Fps is (10 sec: 20070.4, 60 sec: 18256.5, 300 sec: 18256.5). Total num frames: 638976. Throughput: 0: 4196.3. Samples: 146872. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
187
+ [2024-10-08 03:44:50,507][04317] Avg episode reward: [(0, '4.530')]
188
+ [2024-10-08 03:44:51,250][07021] Updated weights for policy 0, policy_version 160 (0.0011)
189
+ [2024-10-08 03:44:53,333][07021] Updated weights for policy 0, policy_version 170 (0.0012)
190
+ [2024-10-08 03:44:55,487][07021] Updated weights for policy 0, policy_version 180 (0.0012)
191
+ [2024-10-08 03:44:55,505][04317] Fps is (10 sec: 19660.7, 60 sec: 18432.0, 300 sec: 18432.0). Total num frames: 737280. Throughput: 0: 4420.9. Samples: 176836. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
192
+ [2024-10-08 03:44:55,507][04317] Avg episode reward: [(0, '4.479')]
193
+ [2024-10-08 03:44:57,496][07021] Updated weights for policy 0, policy_version 190 (0.0011)
194
+ [2024-10-08 03:44:59,493][07021] Updated weights for policy 0, policy_version 200 (0.0011)
195
+ [2024-10-08 03:45:00,505][04317] Fps is (10 sec: 19660.8, 60 sec: 18568.6, 300 sec: 18568.6). Total num frames: 835584. Throughput: 0: 4598.9. Samples: 206950. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
196
+ [2024-10-08 03:45:00,507][04317] Avg episode reward: [(0, '4.344')]
197
+ [2024-10-08 03:45:01,606][07021] Updated weights for policy 0, policy_version 210 (0.0011)
198
+ [2024-10-08 03:45:03,666][07021] Updated weights for policy 0, policy_version 220 (0.0011)
199
+ [2024-10-08 03:45:05,505][04317] Fps is (10 sec: 20070.3, 60 sec: 18759.7, 300 sec: 18759.7). Total num frames: 937984. Throughput: 0: 4857.5. Samples: 221550. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
200
+ [2024-10-08 03:45:05,508][04317] Avg episode reward: [(0, '4.417')]
201
+ [2024-10-08 03:45:05,651][07021] Updated weights for policy 0, policy_version 230 (0.0011)
202
+ [2024-10-08 03:45:07,624][07021] Updated weights for policy 0, policy_version 240 (0.0011)
203
+ [2024-10-08 03:45:09,616][07021] Updated weights for policy 0, policy_version 250 (0.0011)
204
+ [2024-10-08 03:45:10,505][04317] Fps is (10 sec: 20480.2, 60 sec: 18916.1, 300 sec: 18916.1). Total num frames: 1040384. Throughput: 0: 5043.5. Samples: 252382. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
205
+ [2024-10-08 03:45:10,507][04317] Avg episode reward: [(0, '4.859')]
206
+ [2024-10-08 03:45:10,510][07008] Saving new best policy, reward=4.859!
207
+ [2024-10-08 03:45:11,606][07021] Updated weights for policy 0, policy_version 260 (0.0011)
208
+ [2024-10-08 03:45:13,636][07021] Updated weights for policy 0, policy_version 270 (0.0012)
209
+ [2024-10-08 03:45:15,505][04317] Fps is (10 sec: 20070.4, 60 sec: 18978.1, 300 sec: 18978.1). Total num frames: 1138688. Throughput: 0: 5030.4. Samples: 282786. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
210
+ [2024-10-08 03:45:15,508][04317] Avg episode reward: [(0, '4.682')]
211
+ [2024-10-08 03:45:15,735][07021] Updated weights for policy 0, policy_version 280 (0.0012)
212
+ [2024-10-08 03:45:17,787][07021] Updated weights for policy 0, policy_version 290 (0.0011)
213
+ [2024-10-08 03:45:19,788][07021] Updated weights for policy 0, policy_version 300 (0.0011)
214
+ [2024-10-08 03:45:20,505][04317] Fps is (10 sec: 20069.9, 60 sec: 20138.6, 300 sec: 19093.6). Total num frames: 1241088. Throughput: 0: 5030.0. Samples: 297656. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
215
+ [2024-10-08 03:45:20,508][04317] Avg episode reward: [(0, '4.792')]
216
+ [2024-10-08 03:45:21,798][07021] Updated weights for policy 0, policy_version 310 (0.0011)
217
+ [2024-10-08 03:45:23,776][07021] Updated weights for policy 0, policy_version 320 (0.0011)
218
+ [2024-10-08 03:45:25,505][04317] Fps is (10 sec: 20480.1, 60 sec: 20138.7, 300 sec: 19192.7). Total num frames: 1343488. Throughput: 0: 5035.3. Samples: 328412. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
219
+ [2024-10-08 03:45:25,507][04317] Avg episode reward: [(0, '4.972')]
220
+ [2024-10-08 03:45:25,515][07008] Saving new best policy, reward=4.972!
221
+ [2024-10-08 03:45:25,823][07021] Updated weights for policy 0, policy_version 330 (0.0012)
222
+ [2024-10-08 03:45:27,865][07021] Updated weights for policy 0, policy_version 340 (0.0012)
223
+ [2024-10-08 03:45:29,985][07021] Updated weights for policy 0, policy_version 350 (0.0011)
224
+ [2024-10-08 03:45:30,505][04317] Fps is (10 sec: 20070.6, 60 sec: 20070.4, 300 sec: 19223.9). Total num frames: 1441792. Throughput: 0: 5016.9. Samples: 358158. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
225
+ [2024-10-08 03:45:30,507][04317] Avg episode reward: [(0, '4.786')]
226
+ [2024-10-08 03:45:32,006][07021] Updated weights for policy 0, policy_version 360 (0.0011)
227
+ [2024-10-08 03:45:34,012][07021] Updated weights for policy 0, policy_version 370 (0.0011)
228
+ [2024-10-08 03:45:35,505][04317] Fps is (10 sec: 20070.4, 60 sec: 20138.7, 300 sec: 19302.4). Total num frames: 1544192. Throughput: 0: 5031.2. Samples: 373278. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
229
+ [2024-10-08 03:45:35,507][04317] Avg episode reward: [(0, '4.970')]
230
+ [2024-10-08 03:45:36,002][07021] Updated weights for policy 0, policy_version 380 (0.0011)
231
+ [2024-10-08 03:45:37,997][07021] Updated weights for policy 0, policy_version 390 (0.0011)
232
+ [2024-10-08 03:45:39,995][07021] Updated weights for policy 0, policy_version 400 (0.0011)
233
+ [2024-10-08 03:45:40,505][04317] Fps is (10 sec: 20480.0, 60 sec: 20138.7, 300 sec: 19371.7). Total num frames: 1646592. Throughput: 0: 5049.9. Samples: 404080. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
234
+ [2024-10-08 03:45:40,507][04317] Avg episode reward: [(0, '5.066')]
235
+ [2024-10-08 03:45:40,509][07008] Saving new best policy, reward=5.066!
236
+ [2024-10-08 03:45:42,046][07021] Updated weights for policy 0, policy_version 410 (0.0011)
237
+ [2024-10-08 03:45:44,122][07021] Updated weights for policy 0, policy_version 420 (0.0011)
238
+ [2024-10-08 03:45:45,505][04317] Fps is (10 sec: 20070.4, 60 sec: 20070.4, 300 sec: 19387.7). Total num frames: 1744896. Throughput: 0: 5047.1. Samples: 434070. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
239
+ [2024-10-08 03:45:45,508][04317] Avg episode reward: [(0, '5.450')]
240
+ [2024-10-08 03:45:45,515][07008] Saving new best policy, reward=5.450!
241
+ [2024-10-08 03:45:46,159][07021] Updated weights for policy 0, policy_version 430 (0.0011)
242
+ [2024-10-08 03:45:48,154][07021] Updated weights for policy 0, policy_version 440 (0.0011)
243
+ [2024-10-08 03:45:50,136][07021] Updated weights for policy 0, policy_version 450 (0.0011)
244
+ [2024-10-08 03:45:50,505][04317] Fps is (10 sec: 20070.6, 60 sec: 20138.7, 300 sec: 19445.2). Total num frames: 1847296. Throughput: 0: 5062.3. Samples: 449354. Policy #0 lag: (min: 0.0, avg: 0.8, max: 1.0)
245
+ [2024-10-08 03:45:50,507][04317] Avg episode reward: [(0, '7.312')]
246
+ [2024-10-08 03:45:50,534][07008] Saving new best policy, reward=7.312!
247
+ [2024-10-08 03:45:52,122][07021] Updated weights for policy 0, policy_version 460 (0.0011)
248
+ [2024-10-08 03:45:54,103][07021] Updated weights for policy 0, policy_version 470 (0.0011)
249
+ [2024-10-08 03:45:55,505][04317] Fps is (10 sec: 20889.5, 60 sec: 20275.2, 300 sec: 19537.9). Total num frames: 1953792. Throughput: 0: 5064.3. Samples: 480278. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
250
+ [2024-10-08 03:45:55,508][04317] Avg episode reward: [(0, '8.058')]
251
+ [2024-10-08 03:45:55,515][07008] Saving new best policy, reward=8.058!
252
+ [2024-10-08 03:45:56,098][07021] Updated weights for policy 0, policy_version 480 (0.0011)
253
+ [2024-10-08 03:45:58,150][07021] Updated weights for policy 0, policy_version 490 (0.0012)
254
+ [2024-10-08 03:46:00,178][07021] Updated weights for policy 0, policy_version 500 (0.0011)
255
+ [2024-10-08 03:46:00,505][04317] Fps is (10 sec: 20479.8, 60 sec: 20275.2, 300 sec: 19543.8). Total num frames: 2052096. Throughput: 0: 5058.8. Samples: 510434. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
256
+ [2024-10-08 03:46:00,508][04317] Avg episode reward: [(0, '8.725')]
257
+ [2024-10-08 03:46:00,510][07008] Saving new best policy, reward=8.725!
258
+ [2024-10-08 03:46:02,167][07021] Updated weights for policy 0, policy_version 510 (0.0011)
259
+ [2024-10-08 03:46:04,129][07021] Updated weights for policy 0, policy_version 520 (0.0011)
260
+ [2024-10-08 03:46:05,505][04317] Fps is (10 sec: 20480.0, 60 sec: 20343.5, 300 sec: 19623.6). Total num frames: 2158592. Throughput: 0: 5072.6. Samples: 525922. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
261
+ [2024-10-08 03:46:05,508][04317] Avg episode reward: [(0, '10.800')]
262
+ [2024-10-08 03:46:05,515][07008] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000527_2158592.pth...
263
+ [2024-10-08 03:46:05,584][07008] Saving new best policy, reward=10.800!
264
+ [2024-10-08 03:46:06,099][07021] Updated weights for policy 0, policy_version 530 (0.0011)
265
+ [2024-10-08 03:46:08,054][07021] Updated weights for policy 0, policy_version 540 (0.0012)
266
+ [2024-10-08 03:46:10,077][07021] Updated weights for policy 0, policy_version 550 (0.0011)
267
+ [2024-10-08 03:46:10,505][04317] Fps is (10 sec: 20889.7, 60 sec: 20343.4, 300 sec: 19660.8). Total num frames: 2260992. Throughput: 0: 5081.2. Samples: 557066. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
268
+ [2024-10-08 03:46:10,508][04317] Avg episode reward: [(0, '10.034')]
269
+ [2024-10-08 03:46:12,195][07021] Updated weights for policy 0, policy_version 560 (0.0011)
270
+ [2024-10-08 03:46:14,225][07021] Updated weights for policy 0, policy_version 570 (0.0011)
271
+ [2024-10-08 03:46:15,505][04317] Fps is (10 sec: 20070.4, 60 sec: 20343.5, 300 sec: 19660.8). Total num frames: 2359296. Throughput: 0: 5086.8. Samples: 587064. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
272
+ [2024-10-08 03:46:15,508][04317] Avg episode reward: [(0, '12.551')]
273
+ [2024-10-08 03:46:15,515][07008] Saving new best policy, reward=12.551!
274
+ [2024-10-08 03:46:16,207][07021] Updated weights for policy 0, policy_version 580 (0.0011)
275
+ [2024-10-08 03:46:18,183][07021] Updated weights for policy 0, policy_version 590 (0.0011)
276
+ [2024-10-08 03:46:20,143][07021] Updated weights for policy 0, policy_version 600 (0.0011)
277
+ [2024-10-08 03:46:20,505][04317] Fps is (10 sec: 20070.4, 60 sec: 20343.5, 300 sec: 19693.6). Total num frames: 2461696. Throughput: 0: 5093.4. Samples: 602482. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
278
+ [2024-10-08 03:46:20,508][04317] Avg episode reward: [(0, '13.357')]
279
+ [2024-10-08 03:46:20,534][07008] Saving new best policy, reward=13.357!
280
+ [2024-10-08 03:46:22,131][07021] Updated weights for policy 0, policy_version 610 (0.0011)
281
+ [2024-10-08 03:46:24,133][07021] Updated weights for policy 0, policy_version 620 (0.0011)
282
+ [2024-10-08 03:46:25,505][04317] Fps is (10 sec: 20480.2, 60 sec: 20343.5, 300 sec: 19723.8). Total num frames: 2564096. Throughput: 0: 5098.9. Samples: 633530. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
283
+ [2024-10-08 03:46:25,510][04317] Avg episode reward: [(0, '14.219')]
284
+ [2024-10-08 03:46:25,519][07008] Saving new best policy, reward=14.219!
285
+ [2024-10-08 03:46:26,237][07021] Updated weights for policy 0, policy_version 630 (0.0011)
286
+ [2024-10-08 03:46:28,266][07021] Updated weights for policy 0, policy_version 640 (0.0011)
287
+ [2024-10-08 03:46:30,247][07021] Updated weights for policy 0, policy_version 650 (0.0011)
288
+ [2024-10-08 03:46:30,505][04317] Fps is (10 sec: 20479.9, 60 sec: 20411.7, 300 sec: 19751.8). Total num frames: 2666496. Throughput: 0: 5101.1. Samples: 663618. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
289
+ [2024-10-08 03:46:30,508][04317] Avg episode reward: [(0, '14.796')]
290
+ [2024-10-08 03:46:30,510][07008] Saving new best policy, reward=14.796!
291
+ [2024-10-08 03:46:32,196][07021] Updated weights for policy 0, policy_version 660 (0.0011)
292
+ [2024-10-08 03:46:34,172][07021] Updated weights for policy 0, policy_version 670 (0.0012)
293
+ [2024-10-08 03:46:35,505][04317] Fps is (10 sec: 20479.9, 60 sec: 20411.7, 300 sec: 19777.8). Total num frames: 2768896. Throughput: 0: 5110.9. Samples: 679344. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
294
+ [2024-10-08 03:46:35,507][04317] Avg episode reward: [(0, '16.917')]
295
+ [2024-10-08 03:46:35,515][07008] Saving new best policy, reward=16.917!
296
+ [2024-10-08 03:46:36,141][07021] Updated weights for policy 0, policy_version 680 (0.0011)
297
+ [2024-10-08 03:46:38,172][07021] Updated weights for policy 0, policy_version 690 (0.0011)
298
+ [2024-10-08 03:46:40,264][07021] Updated weights for policy 0, policy_version 700 (0.0011)
299
+ [2024-10-08 03:46:40,505][04317] Fps is (10 sec: 20480.0, 60 sec: 20411.7, 300 sec: 19802.0). Total num frames: 2871296. Throughput: 0: 5107.3. Samples: 710108. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
300
+ [2024-10-08 03:46:40,508][04317] Avg episode reward: [(0, '17.158')]
301
+ [2024-10-08 03:46:40,510][07008] Saving new best policy, reward=17.158!
302
+ [2024-10-08 03:46:42,273][07021] Updated weights for policy 0, policy_version 710 (0.0011)
303
+ [2024-10-08 03:46:44,254][07021] Updated weights for policy 0, policy_version 720 (0.0011)
304
+ [2024-10-08 03:46:45,505][04317] Fps is (10 sec: 20480.1, 60 sec: 20480.0, 300 sec: 19824.7). Total num frames: 2973696. Throughput: 0: 5114.0. Samples: 740564. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
305
+ [2024-10-08 03:46:45,507][04317] Avg episode reward: [(0, '17.717')]
306
+ [2024-10-08 03:46:45,516][07008] Saving new best policy, reward=17.717!
307
+ [2024-10-08 03:46:46,237][07021] Updated weights for policy 0, policy_version 730 (0.0011)
308
+ [2024-10-08 03:46:48,181][07021] Updated weights for policy 0, policy_version 740 (0.0011)
309
+ [2024-10-08 03:46:50,161][07021] Updated weights for policy 0, policy_version 750 (0.0011)
310
+ [2024-10-08 03:46:50,505][04317] Fps is (10 sec: 20480.0, 60 sec: 20480.0, 300 sec: 19845.8). Total num frames: 3076096. Throughput: 0: 5115.9. Samples: 756138. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
311
+ [2024-10-08 03:46:50,508][04317] Avg episode reward: [(0, '17.503')]
312
+ [2024-10-08 03:46:52,150][07021] Updated weights for policy 0, policy_version 760 (0.0011)
313
+ [2024-10-08 03:46:54,257][07021] Updated weights for policy 0, policy_version 770 (0.0012)
314
+ [2024-10-08 03:46:55,505][04317] Fps is (10 sec: 20070.2, 60 sec: 20343.5, 300 sec: 19840.0). Total num frames: 3174400. Throughput: 0: 5103.7. Samples: 786734. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
315
+ [2024-10-08 03:46:55,507][04317] Avg episode reward: [(0, '20.499')]
316
+ [2024-10-08 03:46:55,515][07008] Saving new best policy, reward=20.499!
317
+ [2024-10-08 03:46:56,298][07021] Updated weights for policy 0, policy_version 780 (0.0011)
318
+ [2024-10-08 03:46:58,263][07021] Updated weights for policy 0, policy_version 790 (0.0012)
319
+ [2024-10-08 03:47:00,228][07021] Updated weights for policy 0, policy_version 800 (0.0011)
320
+ [2024-10-08 03:47:00,505][04317] Fps is (10 sec: 20480.0, 60 sec: 20480.0, 300 sec: 19884.2). Total num frames: 3280896. Throughput: 0: 5120.5. Samples: 817488. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
321
+ [2024-10-08 03:47:00,508][04317] Avg episode reward: [(0, '20.218')]
322
+ [2024-10-08 03:47:02,224][07021] Updated weights for policy 0, policy_version 810 (0.0011)
323
+ [2024-10-08 03:47:04,180][07021] Updated weights for policy 0, policy_version 820 (0.0011)
324
+ [2024-10-08 03:47:05,505][04317] Fps is (10 sec: 20889.7, 60 sec: 20411.7, 300 sec: 19901.7). Total num frames: 3383296. Throughput: 0: 5123.0. Samples: 833016. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0)
325
+ [2024-10-08 03:47:05,507][04317] Avg episode reward: [(0, '17.323')]
326
+ [2024-10-08 03:47:06,164][07021] Updated weights for policy 0, policy_version 830 (0.0011)
327
+ [2024-10-08 03:47:08,220][07021] Updated weights for policy 0, policy_version 840 (0.0011)
328
+ [2024-10-08 03:47:10,279][07021] Updated weights for policy 0, policy_version 850 (0.0011)
329
+ [2024-10-08 03:47:10,505][04317] Fps is (10 sec: 20479.9, 60 sec: 20411.7, 300 sec: 19918.3). Total num frames: 3485696. Throughput: 0: 5108.6. Samples: 863418. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
330
+ [2024-10-08 03:47:10,507][04317] Avg episode reward: [(0, '21.221')]
331
+ [2024-10-08 03:47:10,510][07008] Saving new best policy, reward=21.221!
332
+ [2024-10-08 03:47:12,259][07021] Updated weights for policy 0, policy_version 860 (0.0011)
333
+ [2024-10-08 03:47:14,225][07021] Updated weights for policy 0, policy_version 870 (0.0011)
334
+ [2024-10-08 03:47:15,505][04317] Fps is (10 sec: 20480.0, 60 sec: 20480.0, 300 sec: 19933.9). Total num frames: 3588096. Throughput: 0: 5129.9. Samples: 894462. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
335
+ [2024-10-08 03:47:15,508][04317] Avg episode reward: [(0, '21.242')]
336
+ [2024-10-08 03:47:15,516][07008] Saving new best policy, reward=21.242!
337
+ [2024-10-08 03:47:16,208][07021] Updated weights for policy 0, policy_version 880 (0.0012)
338
+ [2024-10-08 03:47:18,209][07021] Updated weights for policy 0, policy_version 890 (0.0011)
339
+ [2024-10-08 03:47:20,203][07021] Updated weights for policy 0, policy_version 900 (0.0011)
340
+ [2024-10-08 03:47:20,505][04317] Fps is (10 sec: 20480.0, 60 sec: 20480.0, 300 sec: 19948.6). Total num frames: 3690496. Throughput: 0: 5124.3. Samples: 909936. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
341
+ [2024-10-08 03:47:20,507][04317] Avg episode reward: [(0, '21.015')]
342
+ [2024-10-08 03:47:22,270][07021] Updated weights for policy 0, policy_version 910 (0.0011)
343
+ [2024-10-08 03:47:24,295][07021] Updated weights for policy 0, policy_version 920 (0.0012)
344
+ [2024-10-08 03:47:25,505][04317] Fps is (10 sec: 20479.9, 60 sec: 20480.0, 300 sec: 19962.6). Total num frames: 3792896. Throughput: 0: 5107.0. Samples: 939922. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
345
+ [2024-10-08 03:47:25,508][04317] Avg episode reward: [(0, '22.854')]
346
+ [2024-10-08 03:47:25,516][07008] Saving new best policy, reward=22.854!
347
+ [2024-10-08 03:47:26,256][07021] Updated weights for policy 0, policy_version 930 (0.0012)
348
+ [2024-10-08 03:47:28,270][07021] Updated weights for policy 0, policy_version 940 (0.0011)
349
+ [2024-10-08 03:47:30,224][07021] Updated weights for policy 0, policy_version 950 (0.0012)
350
+ [2024-10-08 03:47:30,505][04317] Fps is (10 sec: 20479.9, 60 sec: 20480.0, 300 sec: 19975.9). Total num frames: 3895296. Throughput: 0: 5121.6. Samples: 971038. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
351
+ [2024-10-08 03:47:30,508][04317] Avg episode reward: [(0, '20.259')]
352
+ [2024-10-08 03:47:32,193][07021] Updated weights for policy 0, policy_version 960 (0.0010)
353
+ [2024-10-08 03:47:34,180][07021] Updated weights for policy 0, policy_version 970 (0.0011)
354
+ [2024-10-08 03:47:35,505][04317] Fps is (10 sec: 20480.2, 60 sec: 20480.0, 300 sec: 19988.5). Total num frames: 3997696. Throughput: 0: 5123.8. Samples: 986708. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
355
+ [2024-10-08 03:47:35,508][04317] Avg episode reward: [(0, '19.869')]
356
+ [2024-10-08 03:47:35,855][07008] Stopping Batcher_0...
357
+ [2024-10-08 03:47:35,855][07008] Loop batcher_evt_loop terminating...
358
+ [2024-10-08 03:47:35,856][07008] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
359
+ [2024-10-08 03:47:35,855][04317] Component Batcher_0 stopped!
360
+ [2024-10-08 03:47:35,871][07021] Weights refcount: 2 0
361
+ [2024-10-08 03:47:35,873][07021] Stopping InferenceWorker_p0-w0...
362
+ [2024-10-08 03:47:35,873][07021] Loop inference_proc0-0_evt_loop terminating...
363
+ [2024-10-08 03:47:35,873][04317] Component InferenceWorker_p0-w0 stopped!
364
+ [2024-10-08 03:47:35,911][07044] Stopping RolloutWorker_w7...
365
+ [2024-10-08 03:47:35,912][07044] Loop rollout_proc7_evt_loop terminating...
366
+ [2024-10-08 03:47:35,912][04317] Component RolloutWorker_w7 stopped!
367
+ [2024-10-08 03:47:35,917][07031] Stopping RolloutWorker_w5...
368
+ [2024-10-08 03:47:35,918][07031] Loop rollout_proc5_evt_loop terminating...
369
+ [2024-10-08 03:47:35,918][07027] Stopping RolloutWorker_w0...
370
+ [2024-10-08 03:47:35,918][07029] Stopping RolloutWorker_w2...
371
+ [2024-10-08 03:47:35,918][07027] Loop rollout_proc0_evt_loop terminating...
372
+ [2024-10-08 03:47:35,918][07029] Loop rollout_proc2_evt_loop terminating...
373
+ [2024-10-08 03:47:35,917][04317] Component RolloutWorker_w5 stopped!
374
+ [2024-10-08 03:47:35,919][07022] Stopping RolloutWorker_w1...
375
+ [2024-10-08 03:47:35,920][07022] Loop rollout_proc1_evt_loop terminating...
376
+ [2024-10-08 03:47:35,919][04317] Component RolloutWorker_w0 stopped!
377
+ [2024-10-08 03:47:35,921][07030] Stopping RolloutWorker_w4...
378
+ [2024-10-08 03:47:35,921][07032] Stopping RolloutWorker_w6...
379
+ [2024-10-08 03:47:35,922][07030] Loop rollout_proc4_evt_loop terminating...
380
+ [2024-10-08 03:47:35,922][07032] Loop rollout_proc6_evt_loop terminating...
381
+ [2024-10-08 03:47:35,923][07028] Stopping RolloutWorker_w3...
382
+ [2024-10-08 03:47:35,921][04317] Component RolloutWorker_w2 stopped!
383
+ [2024-10-08 03:47:35,923][07028] Loop rollout_proc3_evt_loop terminating...
384
+ [2024-10-08 03:47:35,923][04317] Component RolloutWorker_w1 stopped!
385
+ [2024-10-08 03:47:35,925][04317] Component RolloutWorker_w4 stopped!
386
+ [2024-10-08 03:47:35,926][04317] Component RolloutWorker_w6 stopped!
387
+ [2024-10-08 03:47:35,928][04317] Component RolloutWorker_w3 stopped!
388
+ [2024-10-08 03:47:35,938][07008] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
389
+ [2024-10-08 03:47:36,053][07008] Stopping LearnerWorker_p0...
390
+ [2024-10-08 03:47:36,054][07008] Loop learner_proc0_evt_loop terminating...
391
+ [2024-10-08 03:47:36,053][04317] Component LearnerWorker_p0 stopped!
392
+ [2024-10-08 03:47:36,056][04317] Waiting for process learner_proc0 to stop...
393
+ [2024-10-08 03:47:36,785][04317] Waiting for process inference_proc0-0 to join...
394
+ [2024-10-08 03:47:36,788][04317] Waiting for process rollout_proc0 to join...
395
+ [2024-10-08 03:47:36,790][04317] Waiting for process rollout_proc1 to join...
396
+ [2024-10-08 03:47:36,792][04317] Waiting for process rollout_proc2 to join...
397
+ [2024-10-08 03:47:36,794][04317] Waiting for process rollout_proc3 to join...
398
+ [2024-10-08 03:47:36,796][04317] Waiting for process rollout_proc4 to join...
399
+ [2024-10-08 03:47:36,798][04317] Waiting for process rollout_proc5 to join...
400
+ [2024-10-08 03:47:36,800][04317] Waiting for process rollout_proc6 to join...
401
+ [2024-10-08 03:47:36,801][04317] Waiting for process rollout_proc7 to join...
402
+ [2024-10-08 03:47:36,803][04317] Batcher 0 profile tree view:
403
+ batching: 15.9929, releasing_batches: 0.0206
404
+ [2024-10-08 03:47:36,804][04317] InferenceWorker_p0-w0 profile tree view:
405
+ wait_policy: 0.0000
406
+ wait_policy_total: 3.7280
407
+ update_model: 3.1113
408
+ weight_update: 0.0011
409
+ one_step: 0.0023
410
+ handle_policy_step: 183.7676
411
+ deserialize: 7.5591, stack: 1.2014, obs_to_device_normalize: 44.3436, forward: 85.3254, send_messages: 13.4925
412
+ prepare_outputs: 23.7374
413
+ to_cpu: 15.2000
414
+ [2024-10-08 03:47:36,807][04317] Learner 0 profile tree view:
415
+ misc: 0.0049, prepare_batch: 8.9619
416
+ train: 19.3725
417
+ epoch_init: 0.0057, minibatch_init: 0.0063, losses_postprocess: 0.3112, kl_divergence: 0.4220, after_optimizer: 1.4131
418
+ calculate_losses: 7.8730
419
+ losses_init: 0.0033, forward_head: 0.9183, bptt_initial: 3.5001, tail: 0.6437, advantages_returns: 0.1728, losses: 1.0377
420
+ bptt: 1.4136
421
+ bptt_forward_core: 1.3582
422
+ update: 8.9947
423
+ clip: 1.1159
424
+ [2024-10-08 03:47:36,808][04317] RolloutWorker_w0 profile tree view:
425
+ wait_for_trajectories: 0.1452, enqueue_policy_requests: 7.6561, env_step: 125.3836, overhead: 6.2686, complete_rollouts: 0.2347
426
+ save_policy_outputs: 10.3628
427
+ split_output_tensors: 3.5845
428
+ [2024-10-08 03:47:36,809][04317] RolloutWorker_w7 profile tree view:
429
+ wait_for_trajectories: 0.1455, enqueue_policy_requests: 7.6012, env_step: 125.4226, overhead: 6.2732, complete_rollouts: 0.2301
430
+ save_policy_outputs: 10.2584
431
+ split_output_tensors: 3.5595
432
+ [2024-10-08 03:47:36,815][04317] Loop Runner_EvtLoop terminating...
433
+ [2024-10-08 03:47:36,816][04317] Runner profile tree view:
434
+ main_loop: 208.3583
435
+ [2024-10-08 03:47:36,817][04317] Collected {0: 4005888}, FPS: 19226.0
436
+ [2024-10-08 03:47:47,529][04317] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
437
+ [2024-10-08 03:47:47,531][04317] Overriding arg 'num_workers' with value 1 passed from command line
438
+ [2024-10-08 03:47:47,532][04317] Adding new argument 'no_render'=True that is not in the saved config file!
439
+ [2024-10-08 03:47:47,535][04317] Adding new argument 'save_video'=True that is not in the saved config file!
440
+ [2024-10-08 03:47:47,536][04317] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
441
+ [2024-10-08 03:47:47,537][04317] Adding new argument 'video_name'=None that is not in the saved config file!
442
+ [2024-10-08 03:47:47,539][04317] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file!
443
+ [2024-10-08 03:47:47,540][04317] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
444
+ [2024-10-08 03:47:47,541][04317] Adding new argument 'push_to_hub'=False that is not in the saved config file!
445
+ [2024-10-08 03:47:47,542][04317] Adding new argument 'hf_repository'=None that is not in the saved config file!
446
+ [2024-10-08 03:47:47,544][04317] Adding new argument 'policy_index'=0 that is not in the saved config file!
447
+ [2024-10-08 03:47:47,546][04317] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
448
+ [2024-10-08 03:47:47,547][04317] Adding new argument 'train_script'=None that is not in the saved config file!
449
+ [2024-10-08 03:47:47,548][04317] Adding new argument 'enjoy_script'=None that is not in the saved config file!
450
+ [2024-10-08 03:47:47,550][04317] Using frameskip 1 and render_action_repeat=4 for evaluation
451
+ [2024-10-08 03:47:47,562][04317] Doom resolution: 160x120, resize resolution: (128, 72)
452
+ [2024-10-08 03:47:47,565][04317] RunningMeanStd input shape: (3, 72, 128)
453
+ [2024-10-08 03:47:47,567][04317] RunningMeanStd input shape: (1,)
454
+ [2024-10-08 03:47:47,581][04317] ConvEncoder: input_channels=3
455
+ [2024-10-08 03:47:47,715][04317] Conv encoder output size: 512
456
+ [2024-10-08 03:47:47,717][04317] Policy head output size: 512
457
+ [2024-10-08 03:47:49,581][04317] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
458
+ [2024-10-08 03:47:50,495][04317] Num frames 100...
459
+ [2024-10-08 03:47:50,611][04317] Num frames 200...
460
+ [2024-10-08 03:47:50,725][04317] Num frames 300...
461
+ [2024-10-08 03:47:50,842][04317] Num frames 400...
462
+ [2024-10-08 03:47:50,955][04317] Num frames 500...
463
+ [2024-10-08 03:47:51,097][04317] Avg episode rewards: #0: 10.760, true rewards: #0: 5.760
464
+ [2024-10-08 03:47:51,098][04317] Avg episode reward: 10.760, avg true_objective: 5.760
465
+ [2024-10-08 03:47:51,129][04317] Num frames 600...
466
+ [2024-10-08 03:47:51,241][04317] Num frames 700...
467
+ [2024-10-08 03:47:51,365][04317] Num frames 800...
468
+ [2024-10-08 03:47:51,480][04317] Num frames 900...
469
+ [2024-10-08 03:47:51,597][04317] Num frames 1000...
470
+ [2024-10-08 03:47:51,711][04317] Num frames 1100...
471
+ [2024-10-08 03:47:51,827][04317] Num frames 1200...
472
+ [2024-10-08 03:47:51,942][04317] Num frames 1300...
473
+ [2024-10-08 03:47:52,056][04317] Num frames 1400...
474
+ [2024-10-08 03:47:52,171][04317] Num frames 1500...
475
+ [2024-10-08 03:47:52,288][04317] Num frames 1600...
476
+ [2024-10-08 03:47:52,406][04317] Num frames 1700...
477
+ [2024-10-08 03:47:52,523][04317] Num frames 1800...
478
+ [2024-10-08 03:47:52,641][04317] Num frames 1900...
479
+ [2024-10-08 03:47:52,736][04317] Avg episode rewards: #0: 21.670, true rewards: #0: 9.670
480
+ [2024-10-08 03:47:52,737][04317] Avg episode reward: 21.670, avg true_objective: 9.670
481
+ [2024-10-08 03:47:52,814][04317] Num frames 2000...
482
+ [2024-10-08 03:47:52,926][04317] Num frames 2100...
483
+ [2024-10-08 03:47:53,083][04317] Avg episode rewards: #0: 15.300, true rewards: #0: 7.300
484
+ [2024-10-08 03:47:53,085][04317] Avg episode reward: 15.300, avg true_objective: 7.300
485
+ [2024-10-08 03:47:53,098][04317] Num frames 2200...
486
+ [2024-10-08 03:47:53,210][04317] Num frames 2300...
487
+ [2024-10-08 03:47:53,327][04317] Num frames 2400...
488
+ [2024-10-08 03:47:53,440][04317] Num frames 2500...
489
+ [2024-10-08 03:47:53,558][04317] Num frames 2600...
490
+ [2024-10-08 03:47:53,674][04317] Num frames 2700...
491
+ [2024-10-08 03:47:53,787][04317] Num frames 2800...
492
+ [2024-10-08 03:47:53,899][04317] Num frames 2900...
493
+ [2024-10-08 03:47:54,016][04317] Num frames 3000...
494
+ [2024-10-08 03:47:54,132][04317] Num frames 3100...
495
+ [2024-10-08 03:47:54,247][04317] Num frames 3200...
496
+ [2024-10-08 03:47:54,364][04317] Num frames 3300...
497
+ [2024-10-08 03:47:54,481][04317] Num frames 3400...
498
+ [2024-10-08 03:47:54,599][04317] Num frames 3500...
499
+ [2024-10-08 03:47:54,714][04317] Num frames 3600...
500
+ [2024-10-08 03:47:54,829][04317] Num frames 3700...
501
+ [2024-10-08 03:47:54,944][04317] Num frames 3800...
502
+ [2024-10-08 03:47:55,062][04317] Num frames 3900...
503
+ [2024-10-08 03:47:55,175][04317] Num frames 4000...
504
+ [2024-10-08 03:47:55,292][04317] Num frames 4100...
505
+ [2024-10-08 03:47:55,407][04317] Num frames 4200...
506
+ [2024-10-08 03:47:55,564][04317] Avg episode rewards: #0: 25.725, true rewards: #0: 10.725
507
+ [2024-10-08 03:47:55,565][04317] Avg episode reward: 25.725, avg true_objective: 10.725
508
+ [2024-10-08 03:47:55,578][04317] Num frames 4300...
509
+ [2024-10-08 03:47:55,690][04317] Num frames 4400...
510
+ [2024-10-08 03:47:55,803][04317] Num frames 4500...
511
+ [2024-10-08 03:47:55,916][04317] Num frames 4600...
512
+ [2024-10-08 03:47:56,032][04317] Num frames 4700...
513
+ [2024-10-08 03:47:56,132][04317] Avg episode rewards: #0: 22.076, true rewards: #0: 9.476
514
+ [2024-10-08 03:47:56,133][04317] Avg episode reward: 22.076, avg true_objective: 9.476
515
+ [2024-10-08 03:47:56,206][04317] Num frames 4800...
516
+ [2024-10-08 03:47:56,323][04317] Num frames 4900...
517
+ [2024-10-08 03:47:56,441][04317] Num frames 5000...
518
+ [2024-10-08 03:47:56,599][04317] Avg episode rewards: #0: 19.150, true rewards: #0: 8.483
519
+ [2024-10-08 03:47:56,600][04317] Avg episode reward: 19.150, avg true_objective: 8.483
520
+ [2024-10-08 03:47:56,615][04317] Num frames 5100...
521
+ [2024-10-08 03:47:56,729][04317] Num frames 5200...
522
+ [2024-10-08 03:47:56,842][04317] Num frames 5300...
523
+ [2024-10-08 03:47:56,955][04317] Num frames 5400...
524
+ [2024-10-08 03:47:57,072][04317] Num frames 5500...
525
+ [2024-10-08 03:47:57,184][04317] Num frames 5600...
526
+ [2024-10-08 03:47:57,316][04317] Avg episode rewards: #0: 17.951, true rewards: #0: 8.094
527
+ [2024-10-08 03:47:57,317][04317] Avg episode reward: 17.951, avg true_objective: 8.094
528
+ [2024-10-08 03:47:57,359][04317] Num frames 5700...
529
+ [2024-10-08 03:47:57,474][04317] Num frames 5800...
530
+ [2024-10-08 03:47:57,589][04317] Num frames 5900...
531
+ [2024-10-08 03:47:57,703][04317] Num frames 6000...
532
+ [2024-10-08 03:47:57,818][04317] Num frames 6100...
533
+ [2024-10-08 03:47:57,933][04317] Num frames 6200...
534
+ [2024-10-08 03:47:58,036][04317] Avg episode rewards: #0: 16.802, true rewards: #0: 7.802
535
+ [2024-10-08 03:47:58,037][04317] Avg episode reward: 16.802, avg true_objective: 7.802
536
+ [2024-10-08 03:47:58,104][04317] Num frames 6300...
537
+ [2024-10-08 03:47:58,220][04317] Num frames 6400...
538
+ [2024-10-08 03:47:58,341][04317] Num frames 6500...
539
+ [2024-10-08 03:47:58,459][04317] Num frames 6600...
540
+ [2024-10-08 03:47:58,575][04317] Num frames 6700...
541
+ [2024-10-08 03:47:58,691][04317] Num frames 6800...
542
+ [2024-10-08 03:47:58,813][04317] Num frames 6900...
543
+ [2024-10-08 03:47:58,939][04317] Num frames 7000...
544
+ [2024-10-08 03:47:59,068][04317] Num frames 7100...
545
+ [2024-10-08 03:47:59,188][04317] Num frames 7200...
546
+ [2024-10-08 03:47:59,309][04317] Num frames 7300...
547
+ [2024-10-08 03:47:59,431][04317] Num frames 7400...
548
+ [2024-10-08 03:47:59,550][04317] Num frames 7500...
549
+ [2024-10-08 03:47:59,667][04317] Num frames 7600...
550
+ [2024-10-08 03:47:59,791][04317] Num frames 7700...
551
+ [2024-10-08 03:47:59,955][04317] Avg episode rewards: #0: 19.323, true rewards: #0: 8.657
552
+ [2024-10-08 03:47:59,957][04317] Avg episode reward: 19.323, avg true_objective: 8.657
553
+ [2024-10-08 03:47:59,970][04317] Num frames 7800...
554
+ [2024-10-08 03:48:00,083][04317] Num frames 7900...
555
+ [2024-10-08 03:48:00,195][04317] Num frames 8000...
556
+ [2024-10-08 03:48:00,311][04317] Num frames 8100...
557
+ [2024-10-08 03:48:00,429][04317] Num frames 8200...
558
+ [2024-10-08 03:48:00,546][04317] Num frames 8300...
559
+ [2024-10-08 03:48:00,662][04317] Num frames 8400...
560
+ [2024-10-08 03:48:00,780][04317] Num frames 8500...
561
+ [2024-10-08 03:48:00,898][04317] Num frames 8600...
562
+ [2024-10-08 03:48:01,015][04317] Num frames 8700...
563
+ [2024-10-08 03:48:01,132][04317] Num frames 8800...
564
+ [2024-10-08 03:48:01,251][04317] Num frames 8900...
565
+ [2024-10-08 03:48:01,376][04317] Num frames 9000...
566
+ [2024-10-08 03:48:01,497][04317] Num frames 9100...
567
+ [2024-10-08 03:48:01,620][04317] Num frames 9200...
568
+ [2024-10-08 03:48:01,745][04317] Num frames 9300...
569
+ [2024-10-08 03:48:01,864][04317] Num frames 9400...
570
+ [2024-10-08 03:48:01,985][04317] Avg episode rewards: #0: 21.455, true rewards: #0: 9.455
571
+ [2024-10-08 03:48:01,986][04317] Avg episode reward: 21.455, avg true_objective: 9.455
572
+ [2024-10-08 03:48:24,521][04317] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
573
+ [2024-10-08 03:50:12,684][04317] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
574
+ [2024-10-08 03:50:12,686][04317] Overriding arg 'num_workers' with value 1 passed from command line
575
+ [2024-10-08 03:50:12,687][04317] Adding new argument 'no_render'=True that is not in the saved config file!
576
+ [2024-10-08 03:50:12,689][04317] Adding new argument 'save_video'=True that is not in the saved config file!
577
+ [2024-10-08 03:50:12,691][04317] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
578
+ [2024-10-08 03:50:12,692][04317] Adding new argument 'video_name'=None that is not in the saved config file!
579
+ [2024-10-08 03:50:12,694][04317] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
580
+ [2024-10-08 03:50:12,695][04317] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
581
+ [2024-10-08 03:50:12,697][04317] Adding new argument 'push_to_hub'=True that is not in the saved config file!
582
+ [2024-10-08 03:50:12,699][04317] Adding new argument 'hf_repository'='EntropicLettuce/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
583
+ [2024-10-08 03:50:12,700][04317] Adding new argument 'policy_index'=0 that is not in the saved config file!
584
+ [2024-10-08 03:50:12,702][04317] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
585
+ [2024-10-08 03:50:12,703][04317] Adding new argument 'train_script'=None that is not in the saved config file!
586
+ [2024-10-08 03:50:12,705][04317] Adding new argument 'enjoy_script'=None that is not in the saved config file!
587
+ [2024-10-08 03:50:12,707][04317] Using frameskip 1 and render_action_repeat=4 for evaluation
588
+ [2024-10-08 03:50:12,714][04317] RunningMeanStd input shape: (3, 72, 128)
589
+ [2024-10-08 03:50:12,716][04317] RunningMeanStd input shape: (1,)
590
+ [2024-10-08 03:50:12,728][04317] ConvEncoder: input_channels=3
591
+ [2024-10-08 03:50:12,764][04317] Conv encoder output size: 512
592
+ [2024-10-08 03:50:12,765][04317] Policy head output size: 512
593
+ [2024-10-08 03:50:12,786][04317] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
594
+ [2024-10-08 03:50:13,263][04317] Num frames 100...
595
+ [2024-10-08 03:50:13,376][04317] Num frames 200...
596
+ [2024-10-08 03:50:13,488][04317] Num frames 300...
597
+ [2024-10-08 03:50:13,599][04317] Num frames 400...
598
+ [2024-10-08 03:50:13,716][04317] Num frames 500...
599
+ [2024-10-08 03:50:13,830][04317] Num frames 600...
600
+ [2024-10-08 03:50:13,942][04317] Num frames 700...
601
+ [2024-10-08 03:50:14,056][04317] Num frames 800...
602
+ [2024-10-08 03:50:14,168][04317] Num frames 900...
603
+ [2024-10-08 03:50:14,285][04317] Num frames 1000...
604
+ [2024-10-08 03:50:14,398][04317] Num frames 1100...
605
+ [2024-10-08 03:50:14,513][04317] Num frames 1200...
606
+ [2024-10-08 03:50:14,631][04317] Num frames 1300...
607
+ [2024-10-08 03:50:14,746][04317] Num frames 1400...
608
+ [2024-10-08 03:50:14,860][04317] Num frames 1500...
609
+ [2024-10-08 03:50:14,977][04317] Num frames 1600...
610
+ [2024-10-08 03:50:15,094][04317] Num frames 1700...
611
+ [2024-10-08 03:50:15,181][04317] Avg episode rewards: #0: 47.279, true rewards: #0: 17.280
612
+ [2024-10-08 03:50:15,182][04317] Avg episode reward: 47.279, avg true_objective: 17.280
613
+ [2024-10-08 03:50:15,265][04317] Num frames 1800...
614
+ [2024-10-08 03:50:15,378][04317] Num frames 1900...
615
+ [2024-10-08 03:50:15,491][04317] Num frames 2000...
616
+ [2024-10-08 03:50:15,603][04317] Num frames 2100...
617
+ [2024-10-08 03:50:15,723][04317] Num frames 2200...
618
+ [2024-10-08 03:50:15,839][04317] Num frames 2300...
619
+ [2024-10-08 03:50:15,952][04317] Num frames 2400...
620
+ [2024-10-08 03:50:16,070][04317] Num frames 2500...
621
+ [2024-10-08 03:50:16,185][04317] Num frames 2600...
622
+ [2024-10-08 03:50:16,301][04317] Num frames 2700...
623
+ [2024-10-08 03:50:16,427][04317] Num frames 2800...
624
+ [2024-10-08 03:50:16,541][04317] Num frames 2900...
625
+ [2024-10-08 03:50:16,653][04317] Num frames 3000...
626
+ [2024-10-08 03:50:16,790][04317] Avg episode rewards: #0: 38.845, true rewards: #0: 15.345
627
+ [2024-10-08 03:50:16,791][04317] Avg episode reward: 38.845, avg true_objective: 15.345
628
+ [2024-10-08 03:50:16,828][04317] Num frames 3100...
629
+ [2024-10-08 03:50:16,944][04317] Num frames 3200...
630
+ [2024-10-08 03:50:17,061][04317] Num frames 3300...
631
+ [2024-10-08 03:50:17,174][04317] Num frames 3400...
632
+ [2024-10-08 03:50:17,289][04317] Num frames 3500...
633
+ [2024-10-08 03:50:17,401][04317] Num frames 3600...
634
+ [2024-10-08 03:50:17,515][04317] Num frames 3700...
635
+ [2024-10-08 03:50:17,630][04317] Num frames 3800...
636
+ [2024-10-08 03:50:17,755][04317] Num frames 3900...
637
+ [2024-10-08 03:50:17,877][04317] Num frames 4000...
638
+ [2024-10-08 03:50:17,991][04317] Num frames 4100...
639
+ [2024-10-08 03:50:18,110][04317] Num frames 4200...
640
+ [2024-10-08 03:50:18,223][04317] Num frames 4300...
641
+ [2024-10-08 03:50:18,335][04317] Num frames 4400...
642
+ [2024-10-08 03:50:18,455][04317] Num frames 4500...
643
+ [2024-10-08 03:50:18,578][04317] Num frames 4600...
644
+ [2024-10-08 03:50:18,694][04317] Num frames 4700...
645
+ [2024-10-08 03:50:18,813][04317] Num frames 4800...
646
+ [2024-10-08 03:50:18,932][04317] Num frames 4900...
647
+ [2024-10-08 03:50:19,089][04317] Avg episode rewards: #0: 42.296, true rewards: #0: 16.630
648
+ [2024-10-08 03:50:19,091][04317] Avg episode reward: 42.296, avg true_objective: 16.630
649
+ [2024-10-08 03:50:19,107][04317] Num frames 5000...
650
+ [2024-10-08 03:50:19,232][04317] Num frames 5100...
651
+ [2024-10-08 03:50:19,352][04317] Num frames 5200...
652
+ [2024-10-08 03:50:19,468][04317] Num frames 5300...
653
+ [2024-10-08 03:50:19,582][04317] Num frames 5400...
654
+ [2024-10-08 03:50:19,697][04317] Num frames 5500...
655
+ [2024-10-08 03:50:19,813][04317] Num frames 5600...
656
+ [2024-10-08 03:50:19,925][04317] Num frames 5700...
657
+ [2024-10-08 03:50:20,051][04317] Num frames 5800...
658
+ [2024-10-08 03:50:20,171][04317] Num frames 5900...
659
+ [2024-10-08 03:50:20,286][04317] Num frames 6000...
660
+ [2024-10-08 03:50:20,404][04317] Num frames 6100...
661
+ [2024-10-08 03:50:20,522][04317] Num frames 6200...
662
+ [2024-10-08 03:50:20,640][04317] Num frames 6300...
663
+ [2024-10-08 03:50:20,754][04317] Num frames 6400...
664
+ [2024-10-08 03:50:20,870][04317] Num frames 6500...
665
+ [2024-10-08 03:50:20,985][04317] Num frames 6600...
666
+ [2024-10-08 03:50:21,102][04317] Num frames 6700...
667
+ [2024-10-08 03:50:21,224][04317] Num frames 6800...
668
+ [2024-10-08 03:50:21,342][04317] Num frames 6900...
669
+ [2024-10-08 03:50:21,460][04317] Num frames 7000...
670
+ [2024-10-08 03:50:21,621][04317] Avg episode rewards: #0: 44.722, true rewards: #0: 17.723
671
+ [2024-10-08 03:50:21,623][04317] Avg episode reward: 44.722, avg true_objective: 17.723
672
+ [2024-10-08 03:50:21,637][04317] Num frames 7100...
673
+ [2024-10-08 03:50:21,750][04317] Num frames 7200...
674
+ [2024-10-08 03:50:21,862][04317] Num frames 7300...
675
+ [2024-10-08 03:50:21,975][04317] Num frames 7400...
676
+ [2024-10-08 03:50:22,089][04317] Num frames 7500...
677
+ [2024-10-08 03:50:22,222][04317] Avg episode rewards: #0: 37.537, true rewards: #0: 15.138
678
+ [2024-10-08 03:50:22,223][04317] Avg episode reward: 37.537, avg true_objective: 15.138
679
+ [2024-10-08 03:50:22,258][04317] Num frames 7600...
680
+ [2024-10-08 03:50:22,376][04317] Num frames 7700...
681
+ [2024-10-08 03:50:22,492][04317] Num frames 7800...
682
+ [2024-10-08 03:50:22,608][04317] Num frames 7900...
683
+ [2024-10-08 03:50:22,724][04317] Num frames 8000...
684
+ [2024-10-08 03:50:22,838][04317] Num frames 8100...
685
+ [2024-10-08 03:50:22,951][04317] Num frames 8200...
686
+ [2024-10-08 03:50:23,068][04317] Num frames 8300...
687
+ [2024-10-08 03:50:23,184][04317] Num frames 8400...
688
+ [2024-10-08 03:50:23,301][04317] Num frames 8500...
689
+ [2024-10-08 03:50:23,418][04317] Num frames 8600...
690
+ [2024-10-08 03:50:23,537][04317] Num frames 8700...
691
+ [2024-10-08 03:50:23,656][04317] Num frames 8800...
692
+ [2024-10-08 03:50:23,774][04317] Num frames 8900...
693
+ [2024-10-08 03:50:23,890][04317] Num frames 9000...
694
+ [2024-10-08 03:50:24,006][04317] Num frames 9100...
695
+ [2024-10-08 03:50:24,125][04317] Num frames 9200...
696
+ [2024-10-08 03:50:24,241][04317] Num frames 9300...
697
+ [2024-10-08 03:50:24,357][04317] Num frames 9400...
698
+ [2024-10-08 03:50:24,474][04317] Num frames 9500...
699
+ [2024-10-08 03:50:24,592][04317] Num frames 9600...
700
+ [2024-10-08 03:50:24,730][04317] Avg episode rewards: #0: 40.448, true rewards: #0: 16.115
701
+ [2024-10-08 03:50:24,731][04317] Avg episode reward: 40.448, avg true_objective: 16.115
702
+ [2024-10-08 03:50:24,770][04317] Num frames 9700...
703
+ [2024-10-08 03:50:24,884][04317] Num frames 9800...
704
+ [2024-10-08 03:50:24,999][04317] Num frames 9900...
705
+ [2024-10-08 03:50:25,115][04317] Num frames 10000...
706
+ [2024-10-08 03:50:25,230][04317] Avg episode rewards: #0: 35.218, true rewards: #0: 14.361
707
+ [2024-10-08 03:50:25,232][04317] Avg episode reward: 35.218, avg true_objective: 14.361
708
+ [2024-10-08 03:50:25,288][04317] Num frames 10100...
709
+ [2024-10-08 03:50:25,404][04317] Num frames 10200...
710
+ [2024-10-08 03:50:25,519][04317] Num frames 10300...
711
+ [2024-10-08 03:50:25,636][04317] Num frames 10400...
712
+ [2024-10-08 03:50:25,754][04317] Num frames 10500...
713
+ [2024-10-08 03:50:25,871][04317] Num frames 10600...
714
+ [2024-10-08 03:50:25,986][04317] Num frames 10700...
715
+ [2024-10-08 03:50:26,101][04317] Num frames 10800...
716
+ [2024-10-08 03:50:26,217][04317] Num frames 10900...
717
+ [2024-10-08 03:50:26,336][04317] Num frames 11000...
718
+ [2024-10-08 03:50:26,442][04317] Avg episode rewards: #0: 33.306, true rewards: #0: 13.806
719
+ [2024-10-08 03:50:26,444][04317] Avg episode reward: 33.306, avg true_objective: 13.806
720
+ [2024-10-08 03:50:26,509][04317] Num frames 11100...
721
+ [2024-10-08 03:50:26,625][04317] Num frames 11200...
722
+ [2024-10-08 03:50:26,740][04317] Num frames 11300...
723
+ [2024-10-08 03:50:26,856][04317] Num frames 11400...
724
+ [2024-10-08 03:50:26,974][04317] Num frames 11500...
725
+ [2024-10-08 03:50:27,088][04317] Num frames 11600...
726
+ [2024-10-08 03:50:27,203][04317] Num frames 11700...
727
+ [2024-10-08 03:50:27,279][04317] Avg episode rewards: #0: 31.018, true rewards: #0: 13.019
728
+ [2024-10-08 03:50:27,280][04317] Avg episode reward: 31.018, avg true_objective: 13.019
729
+ [2024-10-08 03:50:27,377][04317] Num frames 11800...
730
+ [2024-10-08 03:50:27,491][04317] Num frames 11900...
731
+ [2024-10-08 03:50:27,607][04317] Num frames 12000...
732
+ [2024-10-08 03:50:27,723][04317] Num frames 12100...
733
+ [2024-10-08 03:50:27,841][04317] Num frames 12200...
734
+ [2024-10-08 03:50:27,957][04317] Num frames 12300...
735
+ [2024-10-08 03:50:28,070][04317] Num frames 12400...
736
+ [2024-10-08 03:50:28,186][04317] Num frames 12500...
737
+ [2024-10-08 03:50:28,303][04317] Num frames 12600...
738
+ [2024-10-08 03:50:28,420][04317] Num frames 12700...
739
+ [2024-10-08 03:50:28,535][04317] Num frames 12800...
740
+ [2024-10-08 03:50:28,652][04317] Num frames 12900...
741
+ [2024-10-08 03:50:28,768][04317] Num frames 13000...
742
+ [2024-10-08 03:50:28,883][04317] Num frames 13100...
743
+ [2024-10-08 03:50:28,997][04317] Num frames 13200...
744
+ [2024-10-08 03:50:29,117][04317] Num frames 13300...
745
+ [2024-10-08 03:50:29,237][04317] Num frames 13400...
746
+ [2024-10-08 03:50:29,358][04317] Num frames 13500...
747
+ [2024-10-08 03:50:29,480][04317] Num frames 13600...
748
+ [2024-10-08 03:50:29,597][04317] Num frames 13700...
749
+ [2024-10-08 03:50:29,717][04317] Num frames 13800...
750
+ [2024-10-08 03:50:29,792][04317] Avg episode rewards: #0: 33.717, true rewards: #0: 13.817
751
+ [2024-10-08 03:50:29,794][04317] Avg episode reward: 33.717, avg true_objective: 13.817
752
+ [2024-10-08 03:51:02,121][04317] Replay video saved to /content/train_dir/default_experiment/replay.mp4!