Hugging Face's logo

moodlep
/
rl_course_vizdoom_health_gathering_supreme

Reinforcement Learning
sample-factory
TensorBoard
deep-reinforcement-learning
Eval Results
Model card Files Metrics Community
moodlep commited on
Commit
2df4012
•
1 Parent(s): aab7002

Upload . with huggingface_hub

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. .summary/0/events.out.tfevents.1677145364.9ce75a2e03cc +3 -0
  3. README.md +56 -0
  4. checkpoint_p0/best_000000843_3452928_reward_29.378.pth +3 -0
  5. checkpoint_p0/checkpoint_000000927_3796992.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 +1055 -0
.gitattributes CHANGED
@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
35
+ replay.mp4 filter=lfs diff=lfs merge=lfs -text
.summary/0/events.out.tfevents.1677145364.9ce75a2e03cc ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebcf6ba05cfc4fe2a351b0cb1e6622ef921e5bef40fe09058b693421e8e290c7
3
+ size 454070
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: 12.12 +/- 5.77
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 moodlep/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_000000843_3452928_reward_29.378.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0fc1e0c50af94670a66583f35f508316c7ca0518cbda1bd3c6d08b22cf220e5
3
+ size 34928614
checkpoint_p0/checkpoint_000000927_3796992.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e9dc3f54b5d51ac4e6cd26c90aff17484a5c3990ecc13461ec8ad5c42ab5657
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:a7c921d644bc44ae1c8966feee82af16d57e5ba77ca7c3db956344e5cd0ab90a
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:ce83fddaf50a57af88fe1351cd41c1666001ab45532f5b9e5fda5f47ff427f8d
3
+ size 23856418
sf_log.txt ADDED
@@ -0,0 +1,1055 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2023-02-23 09:42:48,897][11985] Saving configuration to /content/train_dir/default_experiment/config.json...
2
+ [2023-02-23 09:42:48,905][11985] Rollout worker 0 uses device cpu
3
+ [2023-02-23 09:42:48,906][11985] Rollout worker 1 uses device cpu
4
+ [2023-02-23 09:42:48,908][11985] Rollout worker 2 uses device cpu
5
+ [2023-02-23 09:42:48,909][11985] Rollout worker 3 uses device cpu
6
+ [2023-02-23 09:42:48,912][11985] Rollout worker 4 uses device cpu
7
+ [2023-02-23 09:42:48,919][11985] Rollout worker 5 uses device cpu
8
+ [2023-02-23 09:42:48,923][11985] Rollout worker 6 uses device cpu
9
+ [2023-02-23 09:42:48,924][11985] Rollout worker 7 uses device cpu
10
+ [2023-02-23 09:42:49,107][11985] Using GPUs [0] for process 0 (actually maps to GPUs [0])
11
+ [2023-02-23 09:42:49,109][11985] InferenceWorker_p0-w0: min num requests: 2
12
+ [2023-02-23 09:42:49,154][11985] Starting all processes...
13
+ [2023-02-23 09:42:49,156][11985] Starting process learner_proc0
14
+ [2023-02-23 09:42:49,248][11985] Starting all processes...
15
+ [2023-02-23 09:42:49,259][11985] Starting process inference_proc0-0
16
+ [2023-02-23 09:42:49,260][11985] Starting process rollout_proc0
17
+ [2023-02-23 09:42:49,261][11985] Starting process rollout_proc1
18
+ [2023-02-23 09:42:49,261][11985] Starting process rollout_proc2
19
+ [2023-02-23 09:42:49,261][11985] Starting process rollout_proc3
20
+ [2023-02-23 09:42:49,261][11985] Starting process rollout_proc4
21
+ [2023-02-23 09:42:49,261][11985] Starting process rollout_proc5
22
+ [2023-02-23 09:42:49,261][11985] Starting process rollout_proc6
23
+ [2023-02-23 09:42:49,261][11985] Starting process rollout_proc7
24
+ [2023-02-23 09:42:54,027][12181] Using GPUs [0] for process 0 (actually maps to GPUs [0])
25
+ [2023-02-23 09:42:54,028][12181] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
26
+ [2023-02-23 09:42:54,488][12201] Worker 1 uses CPU cores [1]
27
+ [2023-02-23 09:42:54,627][12195] Using GPUs [0] for process 0 (actually maps to GPUs [0])
28
+ [2023-02-23 09:42:54,627][12195] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
29
+ [2023-02-23 09:42:54,638][12203] Worker 3 uses CPU cores [3]
30
+ [2023-02-23 09:42:54,878][12196] Worker 0 uses CPU cores [0]
31
+ [2023-02-23 09:42:54,935][12204] Worker 4 uses CPU cores [0]
32
+ [2023-02-23 09:42:54,984][12208] Worker 7 uses CPU cores [3]
33
+ [2023-02-23 09:42:55,013][12206] Worker 6 uses CPU cores [2]
34
+ [2023-02-23 09:42:55,026][12181] Num visible devices: 1
35
+ [2023-02-23 09:42:55,029][12195] Num visible devices: 1
36
+ [2023-02-23 09:42:55,048][12181] Starting seed is not provided
37
+ [2023-02-23 09:42:55,048][12181] Using GPUs [0] for process 0 (actually maps to GPUs [0])
38
+ [2023-02-23 09:42:55,048][12181] Initializing actor-critic model on device cuda:0
39
+ [2023-02-23 09:42:55,049][12181] RunningMeanStd input shape: (3, 72, 128)
40
+ [2023-02-23 09:42:55,050][12181] RunningMeanStd input shape: (1,)
41
+ [2023-02-23 09:42:55,062][12205] Worker 5 uses CPU cores [1]
42
+ [2023-02-23 09:42:55,063][12181] ConvEncoder: input_channels=3
43
+ [2023-02-23 09:42:55,066][12202] Worker 2 uses CPU cores [2]
44
+ [2023-02-23 09:42:55,348][12181] Conv encoder output size: 512
45
+ [2023-02-23 09:42:55,349][12181] Policy head output size: 512
46
+ [2023-02-23 09:42:55,394][12181] Created Actor Critic model with architecture:
47
+ [2023-02-23 09:42:55,394][12181] 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
+ [2023-02-23 09:43:02,976][12181] Using optimizer <class 'torch.optim.adam.Adam'>
89
+ [2023-02-23 09:43:02,976][12181] No checkpoints found
90
+ [2023-02-23 09:43:02,977][12181] Did not load from checkpoint, starting from scratch!
91
+ [2023-02-23 09:43:02,977][12181] Initialized policy 0 weights for model version 0
92
+ [2023-02-23 09:43:02,979][12181] Using GPUs [0] for process 0 (actually maps to GPUs [0])
93
+ [2023-02-23 09:43:02,981][12181] LearnerWorker_p0 finished initialization!
94
+ [2023-02-23 09:43:03,080][12195] RunningMeanStd input shape: (3, 72, 128)
95
+ [2023-02-23 09:43:03,082][12195] RunningMeanStd input shape: (1,)
96
+ [2023-02-23 09:43:03,098][12195] ConvEncoder: input_channels=3
97
+ [2023-02-23 09:43:03,199][12195] Conv encoder output size: 512
98
+ [2023-02-23 09:43:03,199][12195] Policy head output size: 512
99
+ [2023-02-23 09:43:04,449][11985] 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)
100
+ [2023-02-23 09:43:05,356][11985] Inference worker 0-0 is ready!
101
+ [2023-02-23 09:43:05,358][11985] All inference workers are ready! Signal rollout workers to start!
102
+ [2023-02-23 09:43:05,406][12206] Doom resolution: 160x120, resize resolution: (128, 72)
103
+ [2023-02-23 09:43:05,409][12196] Doom resolution: 160x120, resize resolution: (128, 72)
104
+ [2023-02-23 09:43:05,409][12202] Doom resolution: 160x120, resize resolution: (128, 72)
105
+ [2023-02-23 09:43:05,409][12201] Doom resolution: 160x120, resize resolution: (128, 72)
106
+ [2023-02-23 09:43:05,410][12205] Doom resolution: 160x120, resize resolution: (128, 72)
107
+ [2023-02-23 09:43:05,411][12204] Doom resolution: 160x120, resize resolution: (128, 72)
108
+ [2023-02-23 09:43:05,408][12208] Doom resolution: 160x120, resize resolution: (128, 72)
109
+ [2023-02-23 09:43:05,417][12203] Doom resolution: 160x120, resize resolution: (128, 72)
110
+ [2023-02-23 09:43:05,842][12196] Decorrelating experience for 0 frames...
111
+ [2023-02-23 09:43:05,842][12206] Decorrelating experience for 0 frames...
112
+ [2023-02-23 09:43:05,890][12201] Decorrelating experience for 0 frames...
113
+ [2023-02-23 09:43:05,895][12205] Decorrelating experience for 0 frames...
114
+ [2023-02-23 09:43:06,179][12206] Decorrelating experience for 32 frames...
115
+ [2023-02-23 09:43:06,304][12203] Decorrelating experience for 0 frames...
116
+ [2023-02-23 09:43:06,465][12196] Decorrelating experience for 32 frames...
117
+ [2023-02-23 09:43:06,464][12208] Decorrelating experience for 0 frames...
118
+ [2023-02-23 09:43:06,568][12205] Decorrelating experience for 32 frames...
119
+ [2023-02-23 09:43:06,614][12204] Decorrelating experience for 0 frames...
120
+ [2023-02-23 09:43:06,709][12206] Decorrelating experience for 64 frames...
121
+ [2023-02-23 09:43:06,738][12203] Decorrelating experience for 32 frames...
122
+ [2023-02-23 09:43:07,174][12204] Decorrelating experience for 32 frames...
123
+ [2023-02-23 09:43:07,198][12202] Decorrelating experience for 0 frames...
124
+ [2023-02-23 09:43:07,332][12205] Decorrelating experience for 64 frames...
125
+ [2023-02-23 09:43:07,338][12206] Decorrelating experience for 96 frames...
126
+ [2023-02-23 09:43:07,386][12201] Decorrelating experience for 32 frames...
127
+ [2023-02-23 09:43:07,388][12196] Decorrelating experience for 64 frames...
128
+ [2023-02-23 09:43:07,461][12208] Decorrelating experience for 32 frames...
129
+ [2023-02-23 09:43:07,517][12203] Decorrelating experience for 64 frames...
130
+ [2023-02-23 09:43:07,771][12202] Decorrelating experience for 32 frames...
131
+ [2023-02-23 09:43:07,890][12204] Decorrelating experience for 64 frames...
132
+ [2023-02-23 09:43:08,094][12205] Decorrelating experience for 96 frames...
133
+ [2023-02-23 09:43:08,226][12201] Decorrelating experience for 64 frames...
134
+ [2023-02-23 09:43:08,279][12203] Decorrelating experience for 96 frames...
135
+ [2023-02-23 09:43:08,298][12202] Decorrelating experience for 64 frames...
136
+ [2023-02-23 09:43:08,300][12208] Decorrelating experience for 64 frames...
137
+ [2023-02-23 09:43:08,542][12201] Decorrelating experience for 96 frames...
138
+ [2023-02-23 09:43:08,597][12196] Decorrelating experience for 96 frames...
139
+ [2023-02-23 09:43:08,679][12204] Decorrelating experience for 96 frames...
140
+ [2023-02-23 09:43:08,775][12202] Decorrelating experience for 96 frames...
141
+ [2023-02-23 09:43:08,813][12208] Decorrelating experience for 96 frames...
142
+ [2023-02-23 09:43:09,087][11985] Heartbeat connected on Batcher_0
143
+ [2023-02-23 09:43:09,092][11985] Heartbeat connected on LearnerWorker_p0
144
+ [2023-02-23 09:43:09,116][11985] Heartbeat connected on RolloutWorker_w0
145
+ [2023-02-23 09:43:09,121][11985] Heartbeat connected on RolloutWorker_w1
146
+ [2023-02-23 09:43:09,127][11985] Heartbeat connected on RolloutWorker_w2
147
+ [2023-02-23 09:43:09,131][11985] Heartbeat connected on RolloutWorker_w3
148
+ [2023-02-23 09:43:09,136][11985] Heartbeat connected on RolloutWorker_w4
149
+ [2023-02-23 09:43:09,140][11985] Heartbeat connected on RolloutWorker_w5
150
+ [2023-02-23 09:43:09,145][11985] Heartbeat connected on RolloutWorker_w6
151
+ [2023-02-23 09:43:09,149][11985] Heartbeat connected on RolloutWorker_w7
152
+ [2023-02-23 09:43:09,449][11985] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
153
+ [2023-02-23 09:43:11,163][11985] Heartbeat connected on InferenceWorker_p0-w0
154
+ [2023-02-23 09:43:12,330][12181] Signal inference workers to stop experience collection...
155
+ [2023-02-23 09:43:12,338][12195] InferenceWorker_p0-w0: stopping experience collection
156
+ [2023-02-23 09:43:14,449][11985] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 236.0. Samples: 2360. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
157
+ [2023-02-23 09:43:14,451][11985] Avg episode reward: [(0, '2.158')]
158
+ [2023-02-23 09:43:15,043][12181] Signal inference workers to resume experience collection...
159
+ [2023-02-23 09:43:15,044][12195] InferenceWorker_p0-w0: resuming experience collection
160
+ [2023-02-23 09:43:19,449][11985] Fps is (10 sec: 3276.7, 60 sec: 2184.5, 300 sec: 2184.5). Total num frames: 32768. Throughput: 0: 355.5. Samples: 5332. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
161
+ [2023-02-23 09:43:19,455][11985] Avg episode reward: [(0, '3.785')]
162
+ [2023-02-23 09:43:20,203][12195] Updated weights for policy 0, policy_version 10 (0.0018)
163
+ [2023-02-23 09:43:24,449][11985] Fps is (10 sec: 7782.3, 60 sec: 3891.2, 300 sec: 3891.2). Total num frames: 77824. Throughput: 0: 897.3. Samples: 17946. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
164
+ [2023-02-23 09:43:24,452][11985] Avg episode reward: [(0, '4.565')]
165
+ [2023-02-23 09:43:24,722][12195] Updated weights for policy 0, policy_version 20 (0.0018)
166
+ [2023-02-23 09:43:28,925][12195] Updated weights for policy 0, policy_version 30 (0.0015)
167
+ [2023-02-23 09:43:29,449][11985] Fps is (10 sec: 9420.9, 60 sec: 5079.0, 300 sec: 5079.0). Total num frames: 126976. Throughput: 0: 1288.5. Samples: 32212. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
168
+ [2023-02-23 09:43:29,455][11985] Avg episode reward: [(0, '4.511')]
169
+ [2023-02-23 09:43:29,458][12181] Saving new best policy, reward=4.511!
170
+ [2023-02-23 09:43:34,348][12195] Updated weights for policy 0, policy_version 40 (0.0021)
171
+ [2023-02-23 09:43:34,449][11985] Fps is (10 sec: 8601.6, 60 sec: 5461.3, 300 sec: 5461.3). Total num frames: 163840. Throughput: 0: 1276.1. Samples: 38282. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0)
172
+ [2023-02-23 09:43:34,452][11985] Avg episode reward: [(0, '4.379')]
173
+ [2023-02-23 09:43:39,449][11985] Fps is (10 sec: 6963.2, 60 sec: 5617.4, 300 sec: 5617.4). Total num frames: 196608. Throughput: 0: 1386.5. Samples: 48528. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
174
+ [2023-02-23 09:43:39,453][11985] Avg episode reward: [(0, '4.551')]
175
+ [2023-02-23 09:43:39,528][12181] Saving new best policy, reward=4.551!
176
+ [2023-02-23 09:43:40,012][12195] Updated weights for policy 0, policy_version 50 (0.0012)
177
+ [2023-02-23 09:43:44,449][11985] Fps is (10 sec: 7782.4, 60 sec: 6041.6, 300 sec: 6041.6). Total num frames: 241664. Throughput: 0: 1543.3. Samples: 61734. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
178
+ [2023-02-23 09:43:44,451][11985] Avg episode reward: [(0, '4.583')]
179
+ [2023-02-23 09:43:44,459][12181] Saving new best policy, reward=4.583!
180
+ [2023-02-23 09:43:44,641][12195] Updated weights for policy 0, policy_version 60 (0.0016)
181
+ [2023-02-23 09:43:48,917][12195] Updated weights for policy 0, policy_version 70 (0.0015)
182
+ [2023-02-23 09:43:49,449][11985] Fps is (10 sec: 9420.6, 60 sec: 6462.5, 300 sec: 6462.5). Total num frames: 290816. Throughput: 0: 1525.8. Samples: 68662. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
183
+ [2023-02-23 09:43:49,452][11985] Avg episode reward: [(0, '4.312')]
184
+ [2023-02-23 09:43:54,449][11985] Fps is (10 sec: 8191.9, 60 sec: 6471.7, 300 sec: 6471.7). Total num frames: 323584. Throughput: 0: 1787.5. Samples: 80440. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
185
+ [2023-02-23 09:43:54,452][11985] Avg episode reward: [(0, '4.541')]
186
+ [2023-02-23 09:43:54,599][12195] Updated weights for policy 0, policy_version 80 (0.0012)
187
+ [2023-02-23 09:43:59,449][11985] Fps is (10 sec: 7372.8, 60 sec: 6628.1, 300 sec: 6628.1). Total num frames: 364544. Throughput: 0: 2003.7. Samples: 92528. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
188
+ [2023-02-23 09:43:59,451][11985] Avg episode reward: [(0, '4.515')]
189
+ [2023-02-23 09:43:59,496][12195] Updated weights for policy 0, policy_version 90 (0.0017)
190
+ [2023-02-23 09:44:03,900][12195] Updated weights for policy 0, policy_version 100 (0.0018)
191
+ [2023-02-23 09:44:04,449][11985] Fps is (10 sec: 9011.3, 60 sec: 6894.9, 300 sec: 6894.9). Total num frames: 413696. Throughput: 0: 2092.9. Samples: 99514. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
192
+ [2023-02-23 09:44:04,452][11985] Avg episode reward: [(0, '4.543')]
193
+ [2023-02-23 09:44:08,509][12195] Updated weights for policy 0, policy_version 110 (0.0023)
194
+ [2023-02-23 09:44:09,449][11985] Fps is (10 sec: 9011.3, 60 sec: 7577.6, 300 sec: 6994.7). Total num frames: 454656. Throughput: 0: 2114.2. Samples: 113084. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
195
+ [2023-02-23 09:44:09,452][11985] Avg episode reward: [(0, '4.431')]
196
+ [2023-02-23 09:44:14,269][12195] Updated weights for policy 0, policy_version 120 (0.0013)
197
+ [2023-02-23 09:44:14,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8192.0, 300 sec: 7021.7). Total num frames: 491520. Throughput: 0: 2036.9. Samples: 123874. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
198
+ [2023-02-23 09:44:14,454][11985] Avg episode reward: [(0, '4.407')]
199
+ [2023-02-23 09:44:18,726][12195] Updated weights for policy 0, policy_version 130 (0.0017)
200
+ [2023-02-23 09:44:19,449][11985] Fps is (10 sec: 8192.1, 60 sec: 8396.8, 300 sec: 7154.3). Total num frames: 536576. Throughput: 0: 2050.7. Samples: 130564. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
201
+ [2023-02-23 09:44:19,452][11985] Avg episode reward: [(0, '4.499')]
202
+ [2023-02-23 09:44:23,056][12195] Updated weights for policy 0, policy_version 140 (0.0022)
203
+ [2023-02-23 09:44:24,449][11985] Fps is (10 sec: 9011.2, 60 sec: 8396.8, 300 sec: 7270.4). Total num frames: 581632. Throughput: 0: 2137.7. Samples: 144726. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
204
+ [2023-02-23 09:44:24,451][11985] Avg episode reward: [(0, '4.272')]
205
+ [2023-02-23 09:44:28,236][12195] Updated weights for policy 0, policy_version 150 (0.0018)
206
+ [2023-02-23 09:44:29,449][11985] Fps is (10 sec: 8192.0, 60 sec: 8192.0, 300 sec: 7276.4). Total num frames: 618496. Throughput: 0: 2098.8. Samples: 156180. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
207
+ [2023-02-23 09:44:29,454][11985] Avg episode reward: [(0, '4.315')]
208
+ [2023-02-23 09:44:33,827][12195] Updated weights for policy 0, policy_version 160 (0.0014)
209
+ [2023-02-23 09:44:34,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8260.3, 300 sec: 7327.3). Total num frames: 659456. Throughput: 0: 2062.5. Samples: 161476. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
210
+ [2023-02-23 09:44:34,452][11985] Avg episode reward: [(0, '4.735')]
211
+ [2023-02-23 09:44:34,458][12181] Saving new best policy, reward=4.735!
212
+ [2023-02-23 09:44:38,376][12195] Updated weights for policy 0, policy_version 170 (0.0022)
213
+ [2023-02-23 09:44:39,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8465.1, 300 sec: 7415.9). Total num frames: 704512. Throughput: 0: 2095.2. Samples: 174722. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
214
+ [2023-02-23 09:44:39,452][11985] Avg episode reward: [(0, '4.493')]
215
+ [2023-02-23 09:44:42,750][12195] Updated weights for policy 0, policy_version 180 (0.0014)
216
+ [2023-02-23 09:44:44,449][11985] Fps is (10 sec: 9011.2, 60 sec: 8465.1, 300 sec: 7495.7). Total num frames: 749568. Throughput: 0: 2125.3. Samples: 188168. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
217
+ [2023-02-23 09:44:44,451][11985] Avg episode reward: [(0, '4.674')]
218
+ [2023-02-23 09:44:44,465][12181] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000183_749568.pth...
219
+ [2023-02-23 09:44:48,389][12195] Updated weights for policy 0, policy_version 190 (0.0019)
220
+ [2023-02-23 09:44:49,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8192.0, 300 sec: 7450.8). Total num frames: 782336. Throughput: 0: 2086.6. Samples: 193412. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
221
+ [2023-02-23 09:44:49,454][11985] Avg episode reward: [(0, '4.532')]
222
+ [2023-02-23 09:44:53,385][12195] Updated weights for policy 0, policy_version 200 (0.0019)
223
+ [2023-02-23 09:44:54,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8396.8, 300 sec: 7521.7). Total num frames: 827392. Throughput: 0: 2052.5. Samples: 205446. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
224
+ [2023-02-23 09:44:54,458][11985] Avg episode reward: [(0, '4.616')]
225
+ [2023-02-23 09:44:57,684][12195] Updated weights for policy 0, policy_version 210 (0.0021)
226
+ [2023-02-23 09:44:59,449][11985] Fps is (10 sec: 9420.7, 60 sec: 8533.3, 300 sec: 7622.1). Total num frames: 876544. Throughput: 0: 2129.0. Samples: 219680. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
227
+ [2023-02-23 09:44:59,452][11985] Avg episode reward: [(0, '4.569')]
228
+ [2023-02-23 09:45:02,307][12195] Updated weights for policy 0, policy_version 220 (0.0019)
229
+ [2023-02-23 09:45:04,449][11985] Fps is (10 sec: 8601.3, 60 sec: 8328.5, 300 sec: 7611.7). Total num frames: 913408. Throughput: 0: 2120.4. Samples: 225982. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
230
+ [2023-02-23 09:45:04,451][11985] Avg episode reward: [(0, '4.635')]
231
+ [2023-02-23 09:45:08,111][12195] Updated weights for policy 0, policy_version 230 (0.0012)
232
+ [2023-02-23 09:45:09,449][11985] Fps is (10 sec: 7372.8, 60 sec: 8260.3, 300 sec: 7602.2). Total num frames: 950272. Throughput: 0: 2042.0. Samples: 236614. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
233
+ [2023-02-23 09:45:09,452][11985] Avg episode reward: [(0, '4.648')]
234
+ [2023-02-23 09:45:12,795][12195] Updated weights for policy 0, policy_version 240 (0.0016)
235
+ [2023-02-23 09:45:14,449][11985] Fps is (10 sec: 8192.3, 60 sec: 8396.8, 300 sec: 7656.4). Total num frames: 995328. Throughput: 0: 2088.3. Samples: 250152. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
236
+ [2023-02-23 09:45:14,451][11985] Avg episode reward: [(0, '4.570')]
237
+ [2023-02-23 09:45:17,063][12195] Updated weights for policy 0, policy_version 250 (0.0011)
238
+ [2023-02-23 09:45:19,449][11985] Fps is (10 sec: 9420.9, 60 sec: 8465.1, 300 sec: 7736.9). Total num frames: 1044480. Throughput: 0: 2127.3. Samples: 257206. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
239
+ [2023-02-23 09:45:19,452][11985] Avg episode reward: [(0, '4.853')]
240
+ [2023-02-23 09:45:19,461][12181] Saving new best policy, reward=4.853!
241
+ [2023-02-23 09:45:22,106][12195] Updated weights for policy 0, policy_version 260 (0.0015)
242
+ [2023-02-23 09:45:24,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8328.5, 300 sec: 7723.9). Total num frames: 1081344. Throughput: 0: 2095.6. Samples: 269026. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0)
243
+ [2023-02-23 09:45:24,452][11985] Avg episode reward: [(0, '4.798')]
244
+ [2023-02-23 09:45:27,692][12195] Updated weights for policy 0, policy_version 270 (0.0017)
245
+ [2023-02-23 09:45:29,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8396.8, 300 sec: 7740.0). Total num frames: 1122304. Throughput: 0: 2066.0. Samples: 281136. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
246
+ [2023-02-23 09:45:29,454][11985] Avg episode reward: [(0, '5.125')]
247
+ [2023-02-23 09:45:29,458][12181] Saving new best policy, reward=5.125!
248
+ [2023-02-23 09:45:32,104][12195] Updated weights for policy 0, policy_version 280 (0.0017)
249
+ [2023-02-23 09:45:34,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8465.1, 300 sec: 7782.4). Total num frames: 1167360. Throughput: 0: 2105.3. Samples: 288152. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
250
+ [2023-02-23 09:45:34,452][11985] Avg episode reward: [(0, '5.136')]
251
+ [2023-02-23 09:45:34,459][12181] Saving new best policy, reward=5.136!
252
+ [2023-02-23 09:45:36,400][12195] Updated weights for policy 0, policy_version 290 (0.0014)
253
+ [2023-02-23 09:45:39,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8396.8, 300 sec: 7795.6). Total num frames: 1208320. Throughput: 0: 2136.4. Samples: 301582. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
254
+ [2023-02-23 09:45:39,457][11985] Avg episode reward: [(0, '5.722')]
255
+ [2023-02-23 09:45:39,461][12181] Saving new best policy, reward=5.722!
256
+ [2023-02-23 09:45:41,934][12195] Updated weights for policy 0, policy_version 300 (0.0013)
257
+ [2023-02-23 09:45:44,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8260.3, 300 sec: 7782.4). Total num frames: 1245184. Throughput: 0: 2055.5. Samples: 312178. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
258
+ [2023-02-23 09:45:44,451][11985] Avg episode reward: [(0, '5.316')]
259
+ [2023-02-23 09:45:47,191][12195] Updated weights for policy 0, policy_version 310 (0.0016)
260
+ [2023-02-23 09:45:49,449][11985] Fps is (10 sec: 8192.0, 60 sec: 8465.1, 300 sec: 7819.6). Total num frames: 1290240. Throughput: 0: 2060.9. Samples: 318724. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
261
+ [2023-02-23 09:45:49,452][11985] Avg episode reward: [(0, '5.515')]
262
+ [2023-02-23 09:45:51,437][12195] Updated weights for policy 0, policy_version 320 (0.0021)
263
+ [2023-02-23 09:45:54,449][11985] Fps is (10 sec: 9011.3, 60 sec: 8465.1, 300 sec: 7854.7). Total num frames: 1335296. Throughput: 0: 2139.2. Samples: 332878. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
264
+ [2023-02-23 09:45:54,451][11985] Avg episode reward: [(0, '5.299')]
265
+ [2023-02-23 09:45:55,929][12195] Updated weights for policy 0, policy_version 330 (0.0016)
266
+ [2023-02-23 09:45:59,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8328.5, 300 sec: 7864.3). Total num frames: 1376256. Throughput: 0: 2103.2. Samples: 344796. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
267
+ [2023-02-23 09:45:59,452][11985] Avg episode reward: [(0, '5.580')]
268
+ [2023-02-23 09:46:01,522][12195] Updated weights for policy 0, policy_version 340 (0.0014)
269
+ [2023-02-23 09:46:04,449][11985] Fps is (10 sec: 7782.3, 60 sec: 8328.6, 300 sec: 7850.7). Total num frames: 1413120. Throughput: 0: 2068.1. Samples: 350270. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
270
+ [2023-02-23 09:46:04,452][11985] Avg episode reward: [(0, '5.824')]
271
+ [2023-02-23 09:46:04,461][12181] Saving new best policy, reward=5.824!
272
+ [2023-02-23 09:46:06,311][12195] Updated weights for policy 0, policy_version 350 (0.0015)
273
+ [2023-02-23 09:46:09,449][11985] Fps is (10 sec: 8601.5, 60 sec: 8533.3, 300 sec: 7904.2). Total num frames: 1462272. Throughput: 0: 2112.4. Samples: 364084. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
274
+ [2023-02-23 09:46:09,452][11985] Avg episode reward: [(0, '5.443')]
275
+ [2023-02-23 09:46:10,588][12195] Updated weights for policy 0, policy_version 360 (0.0012)
276
+ [2023-02-23 09:46:14,449][11985] Fps is (10 sec: 9420.9, 60 sec: 8533.3, 300 sec: 7933.3). Total num frames: 1507328. Throughput: 0: 2148.3. Samples: 377810. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
277
+ [2023-02-23 09:46:14,455][11985] Avg episode reward: [(0, '5.873')]
278
+ [2023-02-23 09:46:14,464][12181] Saving new best policy, reward=5.873!
279
+ [2023-02-23 09:46:15,362][12195] Updated weights for policy 0, policy_version 370 (0.0012)
280
+ [2023-02-23 09:46:19,450][11985] Fps is (10 sec: 8191.6, 60 sec: 8328.4, 300 sec: 7918.9). Total num frames: 1544192. Throughput: 0: 2109.6. Samples: 383086. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
281
+ [2023-02-23 09:46:19,453][11985] Avg episode reward: [(0, '5.740')]
282
+ [2023-02-23 09:46:20,916][12195] Updated weights for policy 0, policy_version 380 (0.0014)
283
+ [2023-02-23 09:46:24,449][11985] Fps is (10 sec: 8191.8, 60 sec: 8465.0, 300 sec: 7946.2). Total num frames: 1589248. Throughput: 0: 2087.3. Samples: 395510. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
284
+ [2023-02-23 09:46:24,451][11985] Avg episode reward: [(0, '6.755')]
285
+ [2023-02-23 09:46:24,466][12181] Saving new best policy, reward=6.755!
286
+ [2023-02-23 09:46:25,310][12195] Updated weights for policy 0, policy_version 390 (0.0013)
287
+ [2023-02-23 09:46:29,449][11985] Fps is (10 sec: 9011.8, 60 sec: 8533.3, 300 sec: 7972.2). Total num frames: 1634304. Throughput: 0: 2166.4. Samples: 409666. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
288
+ [2023-02-23 09:46:29,451][11985] Avg episode reward: [(0, '7.916')]
289
+ [2023-02-23 09:46:29,462][12181] Saving new best policy, reward=7.916!
290
+ [2023-02-23 09:46:29,693][12195] Updated weights for policy 0, policy_version 400 (0.0020)
291
+ [2023-02-23 09:46:34,449][11985] Fps is (10 sec: 8601.7, 60 sec: 8465.1, 300 sec: 7977.4). Total num frames: 1675264. Throughput: 0: 2166.8. Samples: 416232. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
292
+ [2023-02-23 09:46:34,452][11985] Avg episode reward: [(0, '7.913')]
293
+ [2023-02-23 09:46:34,768][12195] Updated weights for policy 0, policy_version 410 (0.0012)
294
+ [2023-02-23 09:46:39,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8396.8, 300 sec: 7963.4). Total num frames: 1712128. Throughput: 0: 2091.8. Samples: 427008. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
295
+ [2023-02-23 09:46:39,452][11985] Avg episode reward: [(0, '7.571')]
296
+ [2023-02-23 09:46:40,104][12195] Updated weights for policy 0, policy_version 420 (0.0021)
297
+ [2023-02-23 09:46:44,449][11985] Fps is (10 sec: 8192.0, 60 sec: 8533.3, 300 sec: 7987.2). Total num frames: 1757184. Throughput: 0: 2131.4. Samples: 440708. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
298
+ [2023-02-23 09:46:44,452][11985] Avg episode reward: [(0, '9.034')]
299
+ [2023-02-23 09:46:44,503][12181] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000430_1761280.pth...
300
+ [2023-02-23 09:46:44,510][12195] Updated weights for policy 0, policy_version 430 (0.0012)
301
+ [2023-02-23 09:46:44,605][12181] Saving new best policy, reward=9.034!
302
+ [2023-02-23 09:46:48,871][12195] Updated weights for policy 0, policy_version 440 (0.0019)
303
+ [2023-02-23 09:46:49,449][11985] Fps is (10 sec: 9420.8, 60 sec: 8601.6, 300 sec: 8028.2). Total num frames: 1806336. Throughput: 0: 2165.1. Samples: 447698. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
304
+ [2023-02-23 09:46:49,452][11985] Avg episode reward: [(0, '9.109')]
305
+ [2023-02-23 09:46:49,455][12181] Saving new best policy, reward=9.109!
306
+ [2023-02-23 09:46:54,449][11985] Fps is (10 sec: 8192.1, 60 sec: 8396.8, 300 sec: 7996.1). Total num frames: 1839104. Throughput: 0: 2118.8. Samples: 459430. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
307
+ [2023-02-23 09:46:54,452][11985] Avg episode reward: [(0, '9.757')]
308
+ [2023-02-23 09:46:54,465][12181] Saving new best policy, reward=9.757!
309
+ [2023-02-23 09:46:54,465][12195] Updated weights for policy 0, policy_version 450 (0.0018)
310
+ [2023-02-23 09:46:59,411][12195] Updated weights for policy 0, policy_version 460 (0.0016)
311
+ [2023-02-23 09:46:59,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8465.1, 300 sec: 8017.7). Total num frames: 1884160. Throughput: 0: 2085.8. Samples: 471672. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
312
+ [2023-02-23 09:46:59,451][11985] Avg episode reward: [(0, '11.222')]
313
+ [2023-02-23 09:46:59,455][12181] Saving new best policy, reward=11.222!
314
+ [2023-02-23 09:47:03,610][12195] Updated weights for policy 0, policy_version 470 (0.0015)
315
+ [2023-02-23 09:47:04,449][11985] Fps is (10 sec: 9011.2, 60 sec: 8601.6, 300 sec: 8038.4). Total num frames: 1929216. Throughput: 0: 2128.4. Samples: 478862. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0)
316
+ [2023-02-23 09:47:04,451][11985] Avg episode reward: [(0, '10.488')]
317
+ [2023-02-23 09:47:08,197][12195] Updated weights for policy 0, policy_version 480 (0.0020)
318
+ [2023-02-23 09:47:09,449][11985] Fps is (10 sec: 9011.0, 60 sec: 8533.3, 300 sec: 8058.2). Total num frames: 1974272. Throughput: 0: 2154.6. Samples: 492468. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
319
+ [2023-02-23 09:47:09,452][11985] Avg episode reward: [(0, '11.818')]
320
+ [2023-02-23 09:47:09,454][12181] Saving new best policy, reward=11.818!
321
+ [2023-02-23 09:47:14,031][12195] Updated weights for policy 0, policy_version 490 (0.0016)
322
+ [2023-02-23 09:47:14,450][11985] Fps is (10 sec: 7781.9, 60 sec: 8328.4, 300 sec: 8028.1). Total num frames: 2007040. Throughput: 0: 2077.1. Samples: 503138. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
323
+ [2023-02-23 09:47:14,455][11985] Avg episode reward: [(0, '12.534')]
324
+ [2023-02-23 09:47:14,467][12181] Saving new best policy, reward=12.534!
325
+ [2023-02-23 09:47:18,603][12195] Updated weights for policy 0, policy_version 500 (0.0012)
326
+ [2023-02-23 09:47:19,449][11985] Fps is (10 sec: 8192.2, 60 sec: 8533.4, 300 sec: 8063.5). Total num frames: 2056192. Throughput: 0: 2078.4. Samples: 509760. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
327
+ [2023-02-23 09:47:19,451][11985] Avg episode reward: [(0, '13.720')]
328
+ [2023-02-23 09:47:19,458][12181] Saving new best policy, reward=13.720!
329
+ [2023-02-23 09:47:22,759][12195] Updated weights for policy 0, policy_version 510 (0.0016)
330
+ [2023-02-23 09:47:24,449][11985] Fps is (10 sec: 9421.4, 60 sec: 8533.4, 300 sec: 8081.7). Total num frames: 2101248. Throughput: 0: 2161.0. Samples: 524252. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
331
+ [2023-02-23 09:47:24,451][11985] Avg episode reward: [(0, '16.638')]
332
+ [2023-02-23 09:47:24,505][12181] Saving new best policy, reward=16.638!
333
+ [2023-02-23 09:47:27,636][12195] Updated weights for policy 0, policy_version 520 (0.0014)
334
+ [2023-02-23 09:47:29,449][11985] Fps is (10 sec: 8601.3, 60 sec: 8465.0, 300 sec: 8083.8). Total num frames: 2142208. Throughput: 0: 2120.6. Samples: 536134. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
335
+ [2023-02-23 09:47:29,452][11985] Avg episode reward: [(0, '17.176')]
336
+ [2023-02-23 09:47:29,458][12181] Saving new best policy, reward=17.176!
337
+ [2023-02-23 09:47:33,238][12195] Updated weights for policy 0, policy_version 530 (0.0015)
338
+ [2023-02-23 09:47:34,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8396.8, 300 sec: 8070.6). Total num frames: 2179072. Throughput: 0: 2087.7. Samples: 541644. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
339
+ [2023-02-23 09:47:34,451][11985] Avg episode reward: [(0, '16.530')]
340
+ [2023-02-23 09:47:37,540][12195] Updated weights for policy 0, policy_version 540 (0.0013)
341
+ [2023-02-23 09:47:39,449][11985] Fps is (10 sec: 8601.9, 60 sec: 8601.6, 300 sec: 8102.6). Total num frames: 2228224. Throughput: 0: 2133.3. Samples: 555430. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
342
+ [2023-02-23 09:47:39,452][11985] Avg episode reward: [(0, '17.124')]
343
+ [2023-02-23 09:47:41,919][12195] Updated weights for policy 0, policy_version 550 (0.0025)
344
+ [2023-02-23 09:47:44,449][11985] Fps is (10 sec: 9420.8, 60 sec: 8601.6, 300 sec: 8118.9). Total num frames: 2273280. Throughput: 0: 2160.5. Samples: 568896. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
345
+ [2023-02-23 09:47:44,455][11985] Avg episode reward: [(0, '20.636')]
346
+ [2023-02-23 09:47:44,463][12181] Saving new best policy, reward=20.636!
347
+ [2023-02-23 09:47:47,342][12195] Updated weights for policy 0, policy_version 560 (0.0019)
348
+ [2023-02-23 09:47:49,453][11985] Fps is (10 sec: 7779.4, 60 sec: 8328.0, 300 sec: 8091.3). Total num frames: 2306048. Throughput: 0: 2115.3. Samples: 574058. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
349
+ [2023-02-23 09:47:49,455][11985] Avg episode reward: [(0, '21.991')]
350
+ [2023-02-23 09:47:49,461][12181] Saving new best policy, reward=21.991!
351
+ [2023-02-23 09:47:52,645][12195] Updated weights for policy 0, policy_version 570 (0.0026)
352
+ [2023-02-23 09:47:54,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8533.3, 300 sec: 8107.3). Total num frames: 2351104. Throughput: 0: 2082.1. Samples: 586162. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
353
+ [2023-02-23 09:47:54,454][11985] Avg episode reward: [(0, '21.684')]
354
+ [2023-02-23 09:47:56,960][12195] Updated weights for policy 0, policy_version 580 (0.0022)
355
+ [2023-02-23 09:47:59,449][11985] Fps is (10 sec: 9014.6, 60 sec: 8533.3, 300 sec: 8122.6). Total num frames: 2396160. Throughput: 0: 2161.1. Samples: 600384. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
356
+ [2023-02-23 09:47:59,451][11985] Avg episode reward: [(0, '20.133')]
357
+ [2023-02-23 09:48:01,287][12195] Updated weights for policy 0, policy_version 590 (0.0027)
358
+ [2023-02-23 09:48:04,457][11985] Fps is (10 sec: 8594.7, 60 sec: 8463.9, 300 sec: 8261.2). Total num frames: 2437120. Throughput: 0: 2155.1. Samples: 606756. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
359
+ [2023-02-23 09:48:04,460][11985] Avg episode reward: [(0, '19.120')]
360
+ [2023-02-23 09:48:07,052][12195] Updated weights for policy 0, policy_version 600 (0.0012)
361
+ [2023-02-23 09:48:09,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8328.6, 300 sec: 8386.4). Total num frames: 2473984. Throughput: 0: 2070.9. Samples: 617442. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
362
+ [2023-02-23 09:48:09,452][11985] Avg episode reward: [(0, '19.500')]
363
+ [2023-02-23 09:48:11,830][12195] Updated weights for policy 0, policy_version 610 (0.0010)
364
+ [2023-02-23 09:48:14,449][11985] Fps is (10 sec: 8198.6, 60 sec: 8533.4, 300 sec: 8428.0). Total num frames: 2519040. Throughput: 0: 2114.1. Samples: 631268. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
365
+ [2023-02-23 09:48:14,453][11985] Avg episode reward: [(0, '19.848')]
366
+ [2023-02-23 09:48:16,219][12195] Updated weights for policy 0, policy_version 620 (0.0012)
367
+ [2023-02-23 09:48:19,449][11985] Fps is (10 sec: 9420.6, 60 sec: 8533.3, 300 sec: 8441.9). Total num frames: 2568192. Throughput: 0: 2149.9. Samples: 638392. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
368
+ [2023-02-23 09:48:19,454][11985] Avg episode reward: [(0, '20.705')]
369
+ [2023-02-23 09:48:20,894][12195] Updated weights for policy 0, policy_version 630 (0.0014)
370
+ [2023-02-23 09:48:24,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8396.8, 300 sec: 8400.3). Total num frames: 2605056. Throughput: 0: 2103.4. Samples: 650084. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
371
+ [2023-02-23 09:48:24,452][11985] Avg episode reward: [(0, '19.771')]
372
+ [2023-02-23 09:48:26,658][12195] Updated weights for policy 0, policy_version 640 (0.0015)
373
+ [2023-02-23 09:48:29,449][11985] Fps is (10 sec: 7782.6, 60 sec: 8396.8, 300 sec: 8414.2). Total num frames: 2646016. Throughput: 0: 2072.8. Samples: 662174. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
374
+ [2023-02-23 09:48:29,454][11985] Avg episode reward: [(0, '20.309')]
375
+ [2023-02-23 09:48:31,096][12195] Updated weights for policy 0, policy_version 650 (0.0019)
376
+ [2023-02-23 09:48:34,449][11985] Fps is (10 sec: 8601.4, 60 sec: 8533.3, 300 sec: 8455.8). Total num frames: 2691072. Throughput: 0: 2115.9. Samples: 669266. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
377
+ [2023-02-23 09:48:34,452][11985] Avg episode reward: [(0, '20.493')]
378
+ [2023-02-23 09:48:35,536][12195] Updated weights for policy 0, policy_version 660 (0.0024)
379
+ [2023-02-23 09:48:39,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8396.8, 300 sec: 8441.9). Total num frames: 2732032. Throughput: 0: 2142.2. Samples: 682562. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
380
+ [2023-02-23 09:48:39,452][11985] Avg episode reward: [(0, '20.372')]
381
+ [2023-02-23 09:48:40,741][12195] Updated weights for policy 0, policy_version 670 (0.0013)
382
+ [2023-02-23 09:48:44,449][11985] Fps is (10 sec: 7782.5, 60 sec: 8260.3, 300 sec: 8400.3). Total num frames: 2768896. Throughput: 0: 2059.5. Samples: 693062. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
383
+ [2023-02-23 09:48:44,452][11985] Avg episode reward: [(0, '20.828')]
384
+ [2023-02-23 09:48:44,460][12181] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000676_2768896.pth...
385
+ [2023-02-23 09:48:44,559][12181] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000183_749568.pth
386
+ [2023-02-23 09:48:46,209][12195] Updated weights for policy 0, policy_version 680 (0.0014)
387
+ [2023-02-23 09:48:49,449][11985] Fps is (10 sec: 8192.0, 60 sec: 8465.6, 300 sec: 8441.9). Total num frames: 2813952. Throughput: 0: 2065.5. Samples: 699688. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
388
+ [2023-02-23 09:48:49,453][11985] Avg episode reward: [(0, '21.928')]
389
+ [2023-02-23 09:48:50,514][12195] Updated weights for policy 0, policy_version 690 (0.0021)
390
+ [2023-02-23 09:48:54,449][11985] Fps is (10 sec: 9011.3, 60 sec: 8465.1, 300 sec: 8455.8). Total num frames: 2859008. Throughput: 0: 2143.3. Samples: 713892. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
391
+ [2023-02-23 09:48:54,452][11985] Avg episode reward: [(0, '23.629')]
392
+ [2023-02-23 09:48:54,501][12181] Saving new best policy, reward=23.629!
393
+ [2023-02-23 09:48:55,004][12195] Updated weights for policy 0, policy_version 700 (0.0017)
394
+ [2023-02-23 09:48:59,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8396.8, 300 sec: 8428.0). Total num frames: 2899968. Throughput: 0: 2095.2. Samples: 725554. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
395
+ [2023-02-23 09:48:59,452][11985] Avg episode reward: [(0, '23.590')]
396
+ [2023-02-23 09:49:00,519][12195] Updated weights for policy 0, policy_version 710 (0.0021)
397
+ [2023-02-23 09:49:04,449][11985] Fps is (10 sec: 8192.0, 60 sec: 8397.9, 300 sec: 8428.0). Total num frames: 2940928. Throughput: 0: 2061.2. Samples: 731146. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
398
+ [2023-02-23 09:49:04,451][11985] Avg episode reward: [(0, '22.421')]
399
+ [2023-02-23 09:49:05,289][12195] Updated weights for policy 0, policy_version 720 (0.0013)
400
+ [2023-02-23 09:49:09,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8533.3, 300 sec: 8455.8). Total num frames: 2985984. Throughput: 0: 2117.2. Samples: 745360. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
401
+ [2023-02-23 09:49:09,452][11985] Avg episode reward: [(0, '22.684')]
402
+ [2023-02-23 09:49:09,634][12195] Updated weights for policy 0, policy_version 730 (0.0012)
403
+ [2023-02-23 09:49:14,394][12195] Updated weights for policy 0, policy_version 740 (0.0015)
404
+ [2023-02-23 09:49:14,453][11985] Fps is (10 sec: 9007.6, 60 sec: 8532.8, 300 sec: 8455.7). Total num frames: 3031040. Throughput: 0: 2134.4. Samples: 758232. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
405
+ [2023-02-23 09:49:14,458][11985] Avg episode reward: [(0, '22.940')]
406
+ [2023-02-23 09:49:19,449][11985] Fps is (10 sec: 7782.3, 60 sec: 8260.3, 300 sec: 8414.2). Total num frames: 3063808. Throughput: 0: 2094.2. Samples: 763504. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
407
+ [2023-02-23 09:49:19,452][11985] Avg episode reward: [(0, '25.597')]
408
+ [2023-02-23 09:49:19,454][12181] Saving new best policy, reward=25.597!
409
+ [2023-02-23 09:49:20,260][12195] Updated weights for policy 0, policy_version 750 (0.0014)
410
+ [2023-02-23 09:49:24,449][11985] Fps is (10 sec: 7785.4, 60 sec: 8396.8, 300 sec: 8441.9). Total num frames: 3108864. Throughput: 0: 2071.9. Samples: 775796. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
411
+ [2023-02-23 09:49:24,455][11985] Avg episode reward: [(0, '25.070')]
412
+ [2023-02-23 09:49:24,654][12195] Updated weights for policy 0, policy_version 760 (0.0024)
413
+ [2023-02-23 09:49:29,028][12195] Updated weights for policy 0, policy_version 770 (0.0019)
414
+ [2023-02-23 09:49:29,449][11985] Fps is (10 sec: 9420.9, 60 sec: 8533.3, 300 sec: 8469.7). Total num frames: 3158016. Throughput: 0: 2154.8. Samples: 790028. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
415
+ [2023-02-23 09:49:29,455][11985] Avg episode reward: [(0, '22.735')]
416
+ [2023-02-23 09:49:34,129][12195] Updated weights for policy 0, policy_version 780 (0.0014)
417
+ [2023-02-23 09:49:34,449][11985] Fps is (10 sec: 8601.7, 60 sec: 8396.8, 300 sec: 8441.9). Total num frames: 3194880. Throughput: 0: 2147.2. Samples: 796314. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
418
+ [2023-02-23 09:49:34,451][11985] Avg episode reward: [(0, '23.693')]
419
+ [2023-02-23 09:49:39,449][11985] Fps is (10 sec: 7372.8, 60 sec: 8328.5, 300 sec: 8414.2). Total num frames: 3231744. Throughput: 0: 2070.2. Samples: 807052. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
420
+ [2023-02-23 09:49:39,451][11985] Avg episode reward: [(0, '25.573')]
421
+ [2023-02-23 09:49:39,571][12195] Updated weights for policy 0, policy_version 790 (0.0018)
422
+ [2023-02-23 09:49:43,817][12195] Updated weights for policy 0, policy_version 800 (0.0016)
423
+ [2023-02-23 09:49:44,449][11985] Fps is (10 sec: 8601.5, 60 sec: 8533.3, 300 sec: 8469.7). Total num frames: 3280896. Throughput: 0: 2123.3. Samples: 821104. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
424
+ [2023-02-23 09:49:44,458][11985] Avg episode reward: [(0, '25.345')]
425
+ [2023-02-23 09:49:48,188][12195] Updated weights for policy 0, policy_version 810 (0.0017)
426
+ [2023-02-23 09:49:49,449][11985] Fps is (10 sec: 9420.8, 60 sec: 8533.3, 300 sec: 8469.7). Total num frames: 3325952. Throughput: 0: 2156.2. Samples: 828176. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
427
+ [2023-02-23 09:49:49,453][11985] Avg episode reward: [(0, '26.339')]
428
+ [2023-02-23 09:49:49,463][12181] Saving new best policy, reward=26.339!
429
+ [2023-02-23 09:49:53,630][12195] Updated weights for policy 0, policy_version 820 (0.0018)
430
+ [2023-02-23 09:49:54,449][11985] Fps is (10 sec: 8192.1, 60 sec: 8396.8, 300 sec: 8428.0). Total num frames: 3362816. Throughput: 0: 2101.7. Samples: 839938. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
431
+ [2023-02-23 09:49:54,452][11985] Avg episode reward: [(0, '28.001')]
432
+ [2023-02-23 09:49:54,466][12181] Saving new best policy, reward=28.001!
433
+ [2023-02-23 09:49:58,602][12195] Updated weights for policy 0, policy_version 830 (0.0017)
434
+ [2023-02-23 09:49:59,449][11985] Fps is (10 sec: 8192.0, 60 sec: 8465.1, 300 sec: 8455.8). Total num frames: 3407872. Throughput: 0: 2094.3. Samples: 852468. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
435
+ [2023-02-23 09:49:59,452][11985] Avg episode reward: [(0, '28.260')]
436
+ [2023-02-23 09:49:59,454][12181] Saving new best policy, reward=28.260!
437
+ [2023-02-23 09:50:02,897][12195] Updated weights for policy 0, policy_version 840 (0.0012)
438
+ [2023-02-23 09:50:04,449][11985] Fps is (10 sec: 9011.1, 60 sec: 8533.3, 300 sec: 8483.6). Total num frames: 3452928. Throughput: 0: 2135.8. Samples: 859614. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
439
+ [2023-02-23 09:50:04,452][11985] Avg episode reward: [(0, '29.378')]
440
+ [2023-02-23 09:50:04,461][12181] Saving new best policy, reward=29.378!
441
+ [2023-02-23 09:50:07,221][12195] Updated weights for policy 0, policy_version 850 (0.0020)
442
+ [2023-02-23 09:50:09,449][11985] Fps is (10 sec: 8601.6, 60 sec: 8465.1, 300 sec: 8469.7). Total num frames: 3493888. Throughput: 0: 2160.4. Samples: 873012. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
443
+ [2023-02-23 09:50:09,452][11985] Avg episode reward: [(0, '28.820')]
444
+ [2023-02-23 09:50:12,931][12195] Updated weights for policy 0, policy_version 860 (0.0025)
445
+ [2023-02-23 09:50:14,449][11985] Fps is (10 sec: 7782.5, 60 sec: 8329.1, 300 sec: 8428.0). Total num frames: 3530752. Throughput: 0: 2090.0. Samples: 884080. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
446
+ [2023-02-23 09:50:14,454][11985] Avg episode reward: [(0, '25.738')]
447
+ [2023-02-23 09:50:17,653][12195] Updated weights for policy 0, policy_version 870 (0.0017)
448
+ [2023-02-23 09:50:19,449][11985] Fps is (10 sec: 8601.4, 60 sec: 8601.6, 300 sec: 8469.7). Total num frames: 3579904. Throughput: 0: 2100.9. Samples: 890856. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
449
+ [2023-02-23 09:50:19,452][11985] Avg episode reward: [(0, '24.015')]
450
+ [2023-02-23 09:50:21,905][12195] Updated weights for policy 0, policy_version 880 (0.0024)
451
+ [2023-02-23 09:50:24,449][11985] Fps is (10 sec: 9830.2, 60 sec: 8669.9, 300 sec: 8497.5). Total num frames: 3629056. Throughput: 0: 2189.2. Samples: 905566. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
452
+ [2023-02-23 09:50:24,452][11985] Avg episode reward: [(0, '25.122')]
453
+ [2023-02-23 09:50:26,457][12195] Updated weights for policy 0, policy_version 890 (0.0012)
454
+ [2023-02-23 09:50:29,455][11985] Fps is (10 sec: 8596.4, 60 sec: 8464.2, 300 sec: 8469.5). Total num frames: 3665920. Throughput: 0: 2133.0. Samples: 917104. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
455
+ [2023-02-23 09:50:29,461][11985] Avg episode reward: [(0, '24.957')]
456
+ [2023-02-23 09:50:32,106][12195] Updated weights for policy 0, policy_version 900 (0.0024)
457
+ [2023-02-23 09:50:34,449][11985] Fps is (10 sec: 7782.3, 60 sec: 8533.3, 300 sec: 8469.7). Total num frames: 3706880. Throughput: 0: 2102.6. Samples: 922792. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
458
+ [2023-02-23 09:50:34,452][11985] Avg episode reward: [(0, '25.811')]
459
+ [2023-02-23 09:50:36,509][12195] Updated weights for policy 0, policy_version 910 (0.0019)
460
+ [2023-02-23 09:50:39,449][11985] Fps is (10 sec: 8607.0, 60 sec: 8669.9, 300 sec: 8497.5). Total num frames: 3751936. Throughput: 0: 2155.0. Samples: 936912. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
461
+ [2023-02-23 09:50:39,452][11985] Avg episode reward: [(0, '25.578')]
462
+ [2023-02-23 09:50:40,826][12195] Updated weights for policy 0, policy_version 920 (0.0011)
463
+ [2023-02-23 09:50:44,449][11985] Fps is (10 sec: 9011.4, 60 sec: 8601.6, 300 sec: 8497.5). Total num frames: 3796992. Throughput: 0: 2170.7. Samples: 950150. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
464
+ [2023-02-23 09:50:44,452][11985] Avg episode reward: [(0, '24.904')]
465
+ [2023-02-23 09:50:44,462][12181] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000927_3796992.pth...
466
+ [2023-02-23 09:50:44,563][12181] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000430_1761280.pth
467
+ [2023-02-23 09:50:46,062][12195] Updated weights for policy 0, policy_version 930 (0.0011)
468
+ [2023-02-23 09:50:49,449][11985] Fps is (10 sec: 7782.4, 60 sec: 8396.8, 300 sec: 8455.8). Total num frames: 3829760. Throughput: 0: 2123.5. Samples: 955170. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
469
+ [2023-02-23 09:50:49,452][11985] Avg episode reward: [(0, '25.615')]
470
+ [2023-02-23 09:50:51,603][12195] Updated weights for policy 0, policy_version 940 (0.0015)
471
+ [2023-02-23 09:50:54,449][11985] Fps is (10 sec: 7782.3, 60 sec: 8533.3, 300 sec: 8469.7). Total num frames: 3874816. Throughput: 0: 2094.1. Samples: 967248. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
472
+ [2023-02-23 09:50:54,451][11985] Avg episode reward: [(0, '27.631')]
473
+ [2023-02-23 09:50:55,995][12195] Updated weights for policy 0, policy_version 950 (0.0022)
474
+ [2023-02-23 09:50:59,449][11985] Fps is (10 sec: 9011.2, 60 sec: 8533.3, 300 sec: 8497.5). Total num frames: 3919872. Throughput: 0: 2166.4. Samples: 981568. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
475
+ [2023-02-23 09:50:59,451][11985] Avg episode reward: [(0, '28.712')]
476
+ [2023-02-23 09:51:00,319][12195] Updated weights for policy 0, policy_version 960 (0.0012)
477
+ [2023-02-23 09:51:04,451][11985] Fps is (10 sec: 8599.8, 60 sec: 8464.8, 300 sec: 8469.6). Total num frames: 3960832. Throughput: 0: 2158.0. Samples: 987970. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
478
+ [2023-02-23 09:51:04,453][11985] Avg episode reward: [(0, '28.934')]
479
+ [2023-02-23 09:51:05,559][12195] Updated weights for policy 0, policy_version 970 (0.0016)
480
+ [2023-02-23 09:51:09,449][11985] Fps is (10 sec: 8191.9, 60 sec: 8465.1, 300 sec: 8455.8). Total num frames: 4001792. Throughput: 0: 2082.8. Samples: 999292. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
481
+ [2023-02-23 09:51:09,452][11985] Avg episode reward: [(0, '27.953')]
482
+ [2023-02-23 09:51:09,666][12181] Stopping Batcher_0...
483
+ [2023-02-23 09:51:09,667][12181] Loop batcher_evt_loop terminating...
484
+ [2023-02-23 09:51:09,673][11985] Component Batcher_0 stopped!
485
+ [2023-02-23 09:51:09,678][12181] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
486
+ [2023-02-23 09:51:09,686][11985] Component RolloutWorker_w1 stopped!
487
+ [2023-02-23 09:51:09,688][11985] Component RolloutWorker_w3 stopped!
488
+ [2023-02-23 09:51:09,686][12203] Stopping RolloutWorker_w3...
489
+ [2023-02-23 09:51:09,690][11985] Component RolloutWorker_w7 stopped!
490
+ [2023-02-23 09:51:09,692][12203] Loop rollout_proc3_evt_loop terminating...
491
+ [2023-02-23 09:51:09,693][12201] Stopping RolloutWorker_w1...
492
+ [2023-02-23 09:51:09,697][12201] Loop rollout_proc1_evt_loop terminating...
493
+ [2023-02-23 09:51:09,687][12208] Stopping RolloutWorker_w7...
494
+ [2023-02-23 09:51:09,699][12208] Loop rollout_proc7_evt_loop terminating...
495
+ [2023-02-23 09:51:09,702][11985] Component RolloutWorker_w5 stopped!
496
+ [2023-02-23 09:51:09,708][12204] Stopping RolloutWorker_w4...
497
+ [2023-02-23 09:51:09,710][12204] Loop rollout_proc4_evt_loop terminating...
498
+ [2023-02-23 09:51:09,711][12195] Weights refcount: 2 0
499
+ [2023-02-23 09:51:09,708][11985] Component RolloutWorker_w4 stopped!
500
+ [2023-02-23 09:51:09,705][12205] Stopping RolloutWorker_w5...
501
+ [2023-02-23 09:51:09,714][12205] Loop rollout_proc5_evt_loop terminating...
502
+ [2023-02-23 09:51:09,715][12196] Stopping RolloutWorker_w0...
503
+ [2023-02-23 09:51:09,716][12196] Loop rollout_proc0_evt_loop terminating...
504
+ [2023-02-23 09:51:09,715][12195] Stopping InferenceWorker_p0-w0...
505
+ [2023-02-23 09:51:09,715][11985] Component RolloutWorker_w0 stopped!
506
+ [2023-02-23 09:51:09,717][12195] Loop inference_proc0-0_evt_loop terminating...
507
+ [2023-02-23 09:51:09,717][11985] Component InferenceWorker_p0-w0 stopped!
508
+ [2023-02-23 09:51:09,728][12206] Stopping RolloutWorker_w6...
509
+ [2023-02-23 09:51:09,729][12206] Loop rollout_proc6_evt_loop terminating...
510
+ [2023-02-23 09:51:09,728][11985] Component RolloutWorker_w6 stopped!
511
+ [2023-02-23 09:51:09,741][12202] Stopping RolloutWorker_w2...
512
+ [2023-02-23 09:51:09,741][12202] Loop rollout_proc2_evt_loop terminating...
513
+ [2023-02-23 09:51:09,741][11985] Component RolloutWorker_w2 stopped!
514
+ [2023-02-23 09:51:09,781][12181] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000676_2768896.pth
515
+ [2023-02-23 09:51:09,788][12181] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
516
+ [2023-02-23 09:51:09,905][12181] Stopping LearnerWorker_p0...
517
+ [2023-02-23 09:51:09,905][11985] Component LearnerWorker_p0 stopped!
518
+ [2023-02-23 09:51:09,907][12181] Loop learner_proc0_evt_loop terminating...
519
+ [2023-02-23 09:51:09,911][11985] Waiting for process learner_proc0 to stop...
520
+ [2023-02-23 09:51:11,333][11985] Waiting for process inference_proc0-0 to join...
521
+ [2023-02-23 09:51:11,336][11985] Waiting for process rollout_proc0 to join...
522
+ [2023-02-23 09:51:11,338][11985] Waiting for process rollout_proc1 to join...
523
+ [2023-02-23 09:51:11,340][11985] Waiting for process rollout_proc2 to join...
524
+ [2023-02-23 09:51:11,342][11985] Waiting for process rollout_proc3 to join...
525
+ [2023-02-23 09:51:11,345][11985] Waiting for process rollout_proc4 to join...
526
+ [2023-02-23 09:51:11,346][11985] Waiting for process rollout_proc5 to join...
527
+ [2023-02-23 09:51:11,348][11985] Waiting for process rollout_proc6 to join...
528
+ [2023-02-23 09:51:11,351][11985] Waiting for process rollout_proc7 to join...
529
+ [2023-02-23 09:51:11,353][11985] Batcher 0 profile tree view:
530
+ batching: 19.3320, releasing_batches: 0.0301
531
+ [2023-02-23 09:51:11,354][11985] InferenceWorker_p0-w0 profile tree view:
532
+ wait_policy: 0.0001
533
+ wait_policy_total: 72.7663
534
+ update_model: 6.4008
535
+ weight_update: 0.0012
536
+ one_step: 0.0028
537
+ handle_policy_step: 376.2500
538
+ deserialize: 14.5737, stack: 2.6149, obs_to_device_normalize: 101.7087, forward: 165.2253, send_messages: 18.9011
539
+ prepare_outputs: 54.3567
540
+ to_cpu: 36.7975
541
+ [2023-02-23 09:51:11,356][11985] Learner 0 profile tree view:
542
+ misc: 0.0058, prepare_batch: 13.4537
543
+ train: 60.3458
544
+ epoch_init: 0.0055, minibatch_init: 0.0060, losses_postprocess: 0.5459, kl_divergence: 0.4910, after_optimizer: 27.1247
545
+ calculate_losses: 20.4763
546
+ losses_init: 0.0036, forward_head: 1.0775, bptt_initial: 14.3494, tail: 0.8022, advantages_returns: 0.2661, losses: 1.9835
547
+ bptt: 1.7620
548
+ bptt_forward_core: 1.6849
549
+ update: 11.2384
550
+ clip: 1.2855
551
+ [2023-02-23 09:51:11,357][11985] RolloutWorker_w0 profile tree view:
552
+ wait_for_trajectories: 0.2543, enqueue_policy_requests: 11.5761, env_step: 310.7096, overhead: 12.3420, complete_rollouts: 2.2334
553
+ save_policy_outputs: 11.3773
554
+ split_output_tensors: 5.6039
555
+ [2023-02-23 09:51:11,359][11985] RolloutWorker_w7 profile tree view:
556
+ wait_for_trajectories: 0.2208, enqueue_policy_requests: 11.6518, env_step: 310.6223, overhead: 12.3694, complete_rollouts: 2.3541
557
+ save_policy_outputs: 11.4765
558
+ split_output_tensors: 5.6588
559
+ [2023-02-23 09:51:11,362][11985] Loop Runner_EvtLoop terminating...
560
+ [2023-02-23 09:51:11,364][11985] Runner profile tree view:
561
+ main_loop: 502.2108
562
+ [2023-02-23 09:51:11,366][11985] Collected {0: 4005888}, FPS: 7976.5
563
+ [2023-02-23 09:56:07,776][11985] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
564
+ [2023-02-23 09:56:07,778][11985] Overriding arg 'num_workers' with value 1 passed from command line
565
+ [2023-02-23 09:56:07,780][11985] Adding new argument 'no_render'=True that is not in the saved config file!
566
+ [2023-02-23 09:56:07,782][11985] Adding new argument 'save_video'=True that is not in the saved config file!
567
+ [2023-02-23 09:56:07,783][11985] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
568
+ [2023-02-23 09:56:07,785][11985] Adding new argument 'video_name'=None that is not in the saved config file!
569
+ [2023-02-23 09:56:07,787][11985] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file!
570
+ [2023-02-23 09:56:07,789][11985] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
571
+ [2023-02-23 09:56:07,791][11985] Adding new argument 'push_to_hub'=False that is not in the saved config file!
572
+ [2023-02-23 09:56:07,793][11985] Adding new argument 'hf_repository'=None that is not in the saved config file!
573
+ [2023-02-23 09:56:07,794][11985] Adding new argument 'policy_index'=0 that is not in the saved config file!
574
+ [2023-02-23 09:56:07,796][11985] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
575
+ [2023-02-23 09:56:07,797][11985] Adding new argument 'train_script'=None that is not in the saved config file!
576
+ [2023-02-23 09:56:07,800][11985] Adding new argument 'enjoy_script'=None that is not in the saved config file!
577
+ [2023-02-23 09:56:07,802][11985] Using frameskip 1 and render_action_repeat=4 for evaluation
578
+ [2023-02-23 09:56:07,818][11985] Doom resolution: 160x120, resize resolution: (128, 72)
579
+ [2023-02-23 09:56:07,821][11985] RunningMeanStd input shape: (3, 72, 128)
580
+ [2023-02-23 09:56:07,824][11985] RunningMeanStd input shape: (1,)
581
+ [2023-02-23 09:56:07,839][11985] ConvEncoder: input_channels=3
582
+ [2023-02-23 09:56:08,547][11985] Conv encoder output size: 512
583
+ [2023-02-23 09:56:08,549][11985] Policy head output size: 512
584
+ [2023-02-23 09:56:10,819][11985] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
585
+ [2023-02-23 09:56:12,057][11985] Num frames 100...
586
+ [2023-02-23 09:56:12,162][11985] Num frames 200...
587
+ [2023-02-23 09:56:12,268][11985] Num frames 300...
588
+ [2023-02-23 09:56:12,369][11985] Num frames 400...
589
+ [2023-02-23 09:56:12,471][11985] Num frames 500...
590
+ [2023-02-23 09:56:12,571][11985] Num frames 600...
591
+ [2023-02-23 09:56:12,647][11985] Avg episode rewards: #0: 14.180, true rewards: #0: 6.180
592
+ [2023-02-23 09:56:12,649][11985] Avg episode reward: 14.180, avg true_objective: 6.180
593
+ [2023-02-23 09:56:12,745][11985] Num frames 700...
594
+ [2023-02-23 09:56:12,844][11985] Num frames 800...
595
+ [2023-02-23 09:56:12,949][11985] Num frames 900...
596
+ [2023-02-23 09:56:13,070][11985] Num frames 1000...
597
+ [2023-02-23 09:56:13,195][11985] Num frames 1100...
598
+ [2023-02-23 09:56:13,307][11985] Num frames 1200...
599
+ [2023-02-23 09:56:13,418][11985] Num frames 1300...
600
+ [2023-02-23 09:56:13,532][11985] Num frames 1400...
601
+ [2023-02-23 09:56:13,660][11985] Num frames 1500...
602
+ [2023-02-23 09:56:13,784][11985] Num frames 1600...
603
+ [2023-02-23 09:56:13,897][11985] Num frames 1700...
604
+ [2023-02-23 09:56:14,009][11985] Num frames 1800...
605
+ [2023-02-23 09:56:14,108][11985] Avg episode rewards: #0: 22.670, true rewards: #0: 9.170
606
+ [2023-02-23 09:56:14,111][11985] Avg episode reward: 22.670, avg true_objective: 9.170
607
+ [2023-02-23 09:56:14,200][11985] Num frames 1900...
608
+ [2023-02-23 09:56:14,338][11985] Num frames 2000...
609
+ [2023-02-23 09:56:14,456][11985] Num frames 2100...
610
+ [2023-02-23 09:56:14,573][11985] Num frames 2200...
611
+ [2023-02-23 09:56:14,695][11985] Num frames 2300...
612
+ [2023-02-23 09:56:14,807][11985] Num frames 2400...
613
+ [2023-02-23 09:56:14,922][11985] Num frames 2500...
614
+ [2023-02-23 09:56:15,037][11985] Num frames 2600...
615
+ [2023-02-23 09:56:15,148][11985] Num frames 2700...
616
+ [2023-02-23 09:56:15,263][11985] Num frames 2800...
617
+ [2023-02-23 09:56:15,376][11985] Num frames 2900...
618
+ [2023-02-23 09:56:15,491][11985] Num frames 3000...
619
+ [2023-02-23 09:56:15,630][11985] Avg episode rewards: #0: 23.607, true rewards: #0: 10.273
620
+ [2023-02-23 09:56:15,632][11985] Avg episode reward: 23.607, avg true_objective: 10.273
621
+ [2023-02-23 09:56:15,656][11985] Num frames 3100...
622
+ [2023-02-23 09:56:15,765][11985] Num frames 3200...
623
+ [2023-02-23 09:56:15,868][11985] Num frames 3300...
624
+ [2023-02-23 09:56:15,971][11985] Num frames 3400...
625
+ [2023-02-23 09:56:16,079][11985] Num frames 3500...
626
+ [2023-02-23 09:56:16,183][11985] Num frames 3600...
627
+ [2023-02-23 09:56:16,284][11985] Num frames 3700...
628
+ [2023-02-23 09:56:16,392][11985] Num frames 3800...
629
+ [2023-02-23 09:56:16,499][11985] Num frames 3900...
630
+ [2023-02-23 09:56:16,606][11985] Num frames 4000...
631
+ [2023-02-23 09:56:16,700][11985] Avg episode rewards: #0: 22.838, true rewards: #0: 10.087
632
+ [2023-02-23 09:56:16,703][11985] Avg episode reward: 22.838, avg true_objective: 10.087
633
+ [2023-02-23 09:56:16,775][11985] Num frames 4100...
634
+ [2023-02-23 09:56:16,889][11985] Num frames 4200...
635
+ [2023-02-23 09:56:16,995][11985] Num frames 4300...
636
+ [2023-02-23 09:56:17,105][11985] Num frames 4400...
637
+ [2023-02-23 09:56:17,213][11985] Num frames 4500...
638
+ [2023-02-23 09:56:17,318][11985] Num frames 4600...
639
+ [2023-02-23 09:56:17,420][11985] Num frames 4700...
640
+ [2023-02-23 09:56:17,527][11985] Num frames 4800...
641
+ [2023-02-23 09:56:17,602][11985] Avg episode rewards: #0: 21.636, true rewards: #0: 9.636
642
+ [2023-02-23 09:56:17,604][11985] Avg episode reward: 21.636, avg true_objective: 9.636
643
+ [2023-02-23 09:56:17,694][11985] Num frames 4900...
644
+ [2023-02-23 09:56:17,807][11985] Num frames 5000...
645
+ [2023-02-23 09:56:17,915][11985] Num frames 5100...
646
+ [2023-02-23 09:56:18,017][11985] Num frames 5200...
647
+ [2023-02-23 09:56:18,126][11985] Num frames 5300...
648
+ [2023-02-23 09:56:18,237][11985] Num frames 5400...
649
+ [2023-02-23 09:56:18,351][11985] Num frames 5500...
650
+ [2023-02-23 09:56:18,457][11985] Num frames 5600...
651
+ [2023-02-23 09:56:18,559][11985] Num frames 5700...
652
+ [2023-02-23 09:56:18,664][11985] Num frames 5800...
653
+ [2023-02-23 09:56:18,775][11985] Num frames 5900...
654
+ [2023-02-23 09:56:18,877][11985] Num frames 6000...
655
+ [2023-02-23 09:56:19,001][11985] Num frames 6100...
656
+ [2023-02-23 09:56:19,110][11985] Num frames 6200...
657
+ [2023-02-23 09:56:19,220][11985] Num frames 6300...
658
+ [2023-02-23 09:56:19,334][11985] Num frames 6400...
659
+ [2023-02-23 09:56:19,448][11985] Num frames 6500...
660
+ [2023-02-23 09:56:19,553][11985] Num frames 6600...
661
+ [2023-02-23 09:56:19,667][11985] Num frames 6700...
662
+ [2023-02-23 09:56:19,795][11985] Num frames 6800...
663
+ [2023-02-23 09:56:19,916][11985] Num frames 6900...
664
+ [2023-02-23 09:56:19,993][11985] Avg episode rewards: #0: 27.697, true rewards: #0: 11.530
665
+ [2023-02-23 09:56:19,996][11985] Avg episode reward: 27.697, avg true_objective: 11.530
666
+ [2023-02-23 09:56:20,096][11985] Num frames 7000...
667
+ [2023-02-23 09:56:20,206][11985] Num frames 7100...
668
+ [2023-02-23 09:56:20,312][11985] Num frames 7200...
669
+ [2023-02-23 09:56:20,421][11985] Num frames 7300...
670
+ [2023-02-23 09:56:20,530][11985] Num frames 7400...
671
+ [2023-02-23 09:56:20,637][11985] Num frames 7500...
672
+ [2023-02-23 09:56:20,745][11985] Num frames 7600...
673
+ [2023-02-23 09:56:20,846][11985] Num frames 7700...
674
+ [2023-02-23 09:56:20,945][11985] Num frames 7800...
675
+ [2023-02-23 09:56:21,047][11985] Num frames 7900...
676
+ [2023-02-23 09:56:21,151][11985] Num frames 8000...
677
+ [2023-02-23 09:56:21,258][11985] Num frames 8100...
678
+ [2023-02-23 09:56:21,367][11985] Num frames 8200...
679
+ [2023-02-23 09:56:21,480][11985] Num frames 8300...
680
+ [2023-02-23 09:56:21,597][11985] Num frames 8400...
681
+ [2023-02-23 09:56:21,707][11985] Num frames 8500...
682
+ [2023-02-23 09:56:21,816][11985] Num frames 8600...
683
+ [2023-02-23 09:56:21,921][11985] Num frames 8700...
684
+ [2023-02-23 09:56:22,027][11985] Num frames 8800...
685
+ [2023-02-23 09:56:22,129][11985] Avg episode rewards: #0: 31.203, true rewards: #0: 12.631
686
+ [2023-02-23 09:56:22,130][11985] Avg episode reward: 31.203, avg true_objective: 12.631
687
+ [2023-02-23 09:56:22,196][11985] Num frames 8900...
688
+ [2023-02-23 09:56:22,302][11985] Num frames 9000...
689
+ [2023-02-23 09:56:22,410][11985] Num frames 9100...
690
+ [2023-02-23 09:56:22,516][11985] Num frames 9200...
691
+ [2023-02-23 09:56:22,619][11985] Num frames 9300...
692
+ [2023-02-23 09:56:22,729][11985] Num frames 9400...
693
+ [2023-02-23 09:56:22,833][11985] Num frames 9500...
694
+ [2023-02-23 09:56:22,938][11985] Num frames 9600...
695
+ [2023-02-23 09:56:23,043][11985] Num frames 9700...
696
+ [2023-02-23 09:56:23,152][11985] Num frames 9800...
697
+ [2023-02-23 09:56:23,253][11985] Num frames 9900...
698
+ [2023-02-23 09:56:23,342][11985] Avg episode rewards: #0: 30.162, true rewards: #0: 12.412
699
+ [2023-02-23 09:56:23,345][11985] Avg episode reward: 30.162, avg true_objective: 12.412
700
+ [2023-02-23 09:56:23,423][11985] Num frames 10000...
701
+ [2023-02-23 09:56:23,529][11985] Num frames 10100...
702
+ [2023-02-23 09:56:23,641][11985] Num frames 10200...
703
+ [2023-02-23 09:56:23,750][11985] Num frames 10300...
704
+ [2023-02-23 09:56:23,853][11985] Num frames 10400...
705
+ [2023-02-23 09:56:23,981][11985] Avg episode rewards: #0: 27.971, true rewards: #0: 11.638
706
+ [2023-02-23 09:56:23,983][11985] Avg episode reward: 27.971, avg true_objective: 11.638
707
+ [2023-02-23 09:56:24,014][11985] Num frames 10500...
708
+ [2023-02-23 09:56:24,119][11985] Num frames 10600...
709
+ [2023-02-23 09:56:24,225][11985] Num frames 10700...
710
+ [2023-02-23 09:56:24,329][11985] Num frames 10800...
711
+ [2023-02-23 09:56:24,435][11985] Num frames 10900...
712
+ [2023-02-23 09:56:24,541][11985] Num frames 11000...
713
+ [2023-02-23 09:56:24,642][11985] Num frames 11100...
714
+ [2023-02-23 09:56:24,746][11985] Num frames 11200...
715
+ [2023-02-23 09:56:24,855][11985] Num frames 11300...
716
+ [2023-02-23 09:56:24,966][11985] Num frames 11400...
717
+ [2023-02-23 09:56:25,057][11985] Avg episode rewards: #0: 27.234, true rewards: #0: 11.434
718
+ [2023-02-23 09:56:25,060][11985] Avg episode reward: 27.234, avg true_objective: 11.434
719
+ [2023-02-23 09:57:03,493][11985] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
720
+ [2023-02-23 10:05:26,470][11985] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
721
+ [2023-02-23 10:05:26,471][11985] Overriding arg 'num_workers' with value 1 passed from command line
722
+ [2023-02-23 10:05:26,473][11985] Adding new argument 'no_render'=True that is not in the saved config file!
723
+ [2023-02-23 10:05:26,474][11985] Adding new argument 'save_video'=True that is not in the saved config file!
724
+ [2023-02-23 10:05:26,477][11985] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
725
+ [2023-02-23 10:05:26,478][11985] Adding new argument 'video_name'=None that is not in the saved config file!
726
+ [2023-02-23 10:05:26,479][11985] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
727
+ [2023-02-23 10:05:26,481][11985] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
728
+ [2023-02-23 10:05:26,483][11985] Adding new argument 'push_to_hub'=True that is not in the saved config file!
729
+ [2023-02-23 10:05:26,485][11985] Adding new argument 'hf_repository'='moodlep/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
730
+ [2023-02-23 10:05:26,486][11985] Adding new argument 'policy_index'=0 that is not in the saved config file!
731
+ [2023-02-23 10:05:26,487][11985] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
732
+ [2023-02-23 10:05:26,488][11985] Adding new argument 'train_script'=None that is not in the saved config file!
733
+ [2023-02-23 10:05:26,492][11985] Adding new argument 'enjoy_script'=None that is not in the saved config file!
734
+ [2023-02-23 10:05:26,494][11985] Using frameskip 1 and render_action_repeat=4 for evaluation
735
+ [2023-02-23 10:05:26,506][11985] RunningMeanStd input shape: (3, 72, 128)
736
+ [2023-02-23 10:05:26,508][11985] RunningMeanStd input shape: (1,)
737
+ [2023-02-23 10:05:26,523][11985] ConvEncoder: input_channels=3
738
+ [2023-02-23 10:05:26,558][11985] Conv encoder output size: 512
739
+ [2023-02-23 10:05:26,560][11985] Policy head output size: 512
740
+ [2023-02-23 10:05:26,577][11985] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
741
+ [2023-02-23 10:05:26,977][11985] Num frames 100...
742
+ [2023-02-23 10:05:27,082][11985] Num frames 200...
743
+ [2023-02-23 10:05:27,197][11985] Avg episode rewards: #0: 2.560, true rewards: #0: 2.560
744
+ [2023-02-23 10:05:27,199][11985] Avg episode reward: 2.560, avg true_objective: 2.560
745
+ [2023-02-23 10:05:27,249][11985] Num frames 300...
746
+ [2023-02-23 10:05:27,355][11985] Num frames 400...
747
+ [2023-02-23 10:05:27,463][11985] Num frames 500...
748
+ [2023-02-23 10:05:27,561][11985] Num frames 600...
749
+ [2023-02-23 10:05:27,658][11985] Num frames 700...
750
+ [2023-02-23 10:05:27,764][11985] Num frames 800...
751
+ [2023-02-23 10:05:27,882][11985] Avg episode rewards: #0: 6.805, true rewards: #0: 4.305
752
+ [2023-02-23 10:05:27,884][11985] Avg episode reward: 6.805, avg true_objective: 4.305
753
+ [2023-02-23 10:05:27,929][11985] Num frames 900...
754
+ [2023-02-23 10:05:28,037][11985] Num frames 1000...
755
+ [2023-02-23 10:05:28,145][11985] Num frames 1100...
756
+ [2023-02-23 10:05:28,255][11985] Num frames 1200...
757
+ [2023-02-23 10:05:28,358][11985] Num frames 1300...
758
+ [2023-02-23 10:05:28,462][11985] Num frames 1400...
759
+ [2023-02-23 10:05:28,565][11985] Num frames 1500...
760
+ [2023-02-23 10:05:28,676][11985] Num frames 1600...
761
+ [2023-02-23 10:05:28,778][11985] Num frames 1700...
762
+ [2023-02-23 10:05:28,883][11985] Num frames 1800...
763
+ [2023-02-23 10:05:28,990][11985] Num frames 1900...
764
+ [2023-02-23 10:05:29,097][11985] Num frames 2000...
765
+ [2023-02-23 10:05:29,205][11985] Num frames 2100...
766
+ [2023-02-23 10:05:29,304][11985] Num frames 2200...
767
+ [2023-02-23 10:05:29,414][11985] Num frames 2300...
768
+ [2023-02-23 10:05:29,522][11985] Num frames 2400...
769
+ [2023-02-23 10:05:29,635][11985] Num frames 2500...
770
+ [2023-02-23 10:05:29,758][11985] Num frames 2600...
771
+ [2023-02-23 10:05:29,885][11985] Num frames 2700...
772
+ [2023-02-23 10:05:29,994][11985] Num frames 2800...
773
+ [2023-02-23 10:05:30,104][11985] Num frames 2900...
774
+ [2023-02-23 10:05:30,227][11985] Avg episode rewards: #0: 25.203, true rewards: #0: 9.870
775
+ [2023-02-23 10:05:30,229][11985] Avg episode reward: 25.203, avg true_objective: 9.870
776
+ [2023-02-23 10:05:30,273][11985] Num frames 3000...
777
+ [2023-02-23 10:05:30,379][11985] Num frames 3100...
778
+ [2023-02-23 10:05:30,482][11985] Num frames 3200...
779
+ [2023-02-23 10:05:30,584][11985] Num frames 3300...
780
+ [2023-02-23 10:05:30,685][11985] Num frames 3400...
781
+ [2023-02-23 10:05:30,790][11985] Num frames 3500...
782
+ [2023-02-23 10:05:30,893][11985] Num frames 3600...
783
+ [2023-02-23 10:05:31,005][11985] Num frames 3700...
784
+ [2023-02-23 10:05:31,107][11985] Num frames 3800...
785
+ [2023-02-23 10:05:31,209][11985] Num frames 3900...
786
+ [2023-02-23 10:05:31,319][11985] Num frames 4000...
787
+ [2023-02-23 10:05:31,420][11985] Num frames 4100...
788
+ [2023-02-23 10:05:31,532][11985] Avg episode rewards: #0: 25.630, true rewards: #0: 10.380
789
+ [2023-02-23 10:05:31,533][11985] Avg episode reward: 25.630, avg true_objective: 10.380
790
+ [2023-02-23 10:05:31,590][11985] Num frames 4200...
791
+ [2023-02-23 10:05:31,703][11985] Num frames 4300...
792
+ [2023-02-23 10:05:31,813][11985] Num frames 4400...
793
+ [2023-02-23 10:05:31,914][11985] Num frames 4500...
794
+ [2023-02-23 10:05:32,025][11985] Num frames 4600...
795
+ [2023-02-23 10:05:32,130][11985] Num frames 4700...
796
+ [2023-02-23 10:05:32,239][11985] Num frames 4800...
797
+ [2023-02-23 10:05:32,342][11985] Num frames 4900...
798
+ [2023-02-23 10:05:32,450][11985] Num frames 5000...
799
+ [2023-02-23 10:05:32,549][11985] Num frames 5100...
800
+ [2023-02-23 10:05:32,656][11985] Num frames 5200...
801
+ [2023-02-23 10:05:32,770][11985] Num frames 5300...
802
+ [2023-02-23 10:05:32,881][11985] Num frames 5400...
803
+ [2023-02-23 10:05:32,994][11985] Num frames 5500...
804
+ [2023-02-23 10:05:33,109][11985] Num frames 5600...
805
+ [2023-02-23 10:05:33,215][11985] Num frames 5700...
806
+ [2023-02-23 10:05:33,323][11985] Num frames 5800...
807
+ [2023-02-23 10:05:33,432][11985] Num frames 5900...
808
+ [2023-02-23 10:05:33,539][11985] Num frames 6000...
809
+ [2023-02-23 10:05:33,648][11985] Num frames 6100...
810
+ [2023-02-23 10:05:33,800][11985] Avg episode rewards: #0: 32.900, true rewards: #0: 12.300
811
+ [2023-02-23 10:05:33,802][11985] Avg episode reward: 32.900, avg true_objective: 12.300
812
+ [2023-02-23 10:05:33,860][11985] Num frames 6200...
813
+ [2023-02-23 10:05:33,969][11985] Num frames 6300...
814
+ [2023-02-23 10:05:34,079][11985] Num frames 6400...
815
+ [2023-02-23 10:05:34,186][11985] Num frames 6500...
816
+ [2023-02-23 10:05:34,291][11985] Num frames 6600...
817
+ [2023-02-23 10:05:34,441][11985] Num frames 6700...
818
+ [2023-02-23 10:05:34,550][11985] Num frames 6800...
819
+ [2023-02-23 10:05:34,659][11985] Num frames 6900...
820
+ [2023-02-23 10:05:34,761][11985] Num frames 7000...
821
+ [2023-02-23 10:05:34,869][11985] Num frames 7100...
822
+ [2023-02-23 10:05:34,997][11985] Num frames 7200...
823
+ [2023-02-23 10:05:35,127][11985] Num frames 7300...
824
+ [2023-02-23 10:05:35,237][11985] Avg episode rewards: #0: 32.903, true rewards: #0: 12.237
825
+ [2023-02-23 10:05:35,238][11985] Avg episode reward: 32.903, avg true_objective: 12.237
826
+ [2023-02-23 10:05:35,304][11985] Num frames 7400...
827
+ [2023-02-23 10:05:35,422][11985] Num frames 7500...
828
+ [2023-02-23 10:05:35,535][11985] Num frames 7600...
829
+ [2023-02-23 10:05:35,655][11985] Num frames 7700...
830
+ [2023-02-23 10:05:35,764][11985] Num frames 7800...
831
+ [2023-02-23 10:05:35,881][11985] Num frames 7900...
832
+ [2023-02-23 10:05:35,992][11985] Num frames 8000...
833
+ [2023-02-23 10:05:36,120][11985] Num frames 8100...
834
+ [2023-02-23 10:05:36,240][11985] Num frames 8200...
835
+ [2023-02-23 10:05:36,358][11985] Num frames 8300...
836
+ [2023-02-23 10:05:36,476][11985] Num frames 8400...
837
+ [2023-02-23 10:05:36,587][11985] Num frames 8500...
838
+ [2023-02-23 10:05:36,698][11985] Num frames 8600...
839
+ [2023-02-23 10:05:36,809][11985] Num frames 8700...
840
+ [2023-02-23 10:05:36,926][11985] Num frames 8800...
841
+ [2023-02-23 10:05:37,046][11985] Num frames 8900...
842
+ [2023-02-23 10:05:37,161][11985] Num frames 9000...
843
+ [2023-02-23 10:05:37,277][11985] Num frames 9100...
844
+ [2023-02-23 10:05:37,387][11985] Num frames 9200...
845
+ [2023-02-23 10:05:37,462][11985] Avg episode rewards: #0: 34.742, true rewards: #0: 13.171
846
+ [2023-02-23 10:05:37,463][11985] Avg episode reward: 34.742, avg true_objective: 13.171
847
+ [2023-02-23 10:05:37,550][11985] Num frames 9300...
848
+ [2023-02-23 10:05:37,657][11985] Num frames 9400...
849
+ [2023-02-23 10:05:37,760][11985] Num frames 9500...
850
+ [2023-02-23 10:05:37,863][11985] Num frames 9600...
851
+ [2023-02-23 10:05:37,968][11985] Num frames 9700...
852
+ [2023-02-23 10:05:38,077][11985] Num frames 9800...
853
+ [2023-02-23 10:05:38,182][11985] Num frames 9900...
854
+ [2023-02-23 10:05:38,283][11985] Num frames 10000...
855
+ [2023-02-23 10:05:38,432][11985] Avg episode rewards: #0: 32.605, true rewards: #0: 12.605
856
+ [2023-02-23 10:05:38,434][11985] Avg episode reward: 32.605, avg true_objective: 12.605
857
+ [2023-02-23 10:05:38,455][11985] Num frames 10100...
858
+ [2023-02-23 10:05:38,561][11985] Num frames 10200...
859
+ [2023-02-23 10:05:38,666][11985] Num frames 10300...
860
+ [2023-02-23 10:05:38,776][11985] Num frames 10400...
861
+ [2023-02-23 10:05:38,877][11985] Num frames 10500...
862
+ [2023-02-23 10:05:38,979][11985] Num frames 10600...
863
+ [2023-02-23 10:05:39,097][11985] Num frames 10700...
864
+ [2023-02-23 10:05:39,209][11985] Num frames 10800...
865
+ [2023-02-23 10:05:39,311][11985] Num frames 10900...
866
+ [2023-02-23 10:05:39,420][11985] Num frames 11000...
867
+ [2023-02-23 10:05:39,529][11985] Num frames 11100...
868
+ [2023-02-23 10:05:39,637][11985] Num frames 11200...
869
+ [2023-02-23 10:05:39,740][11985] Num frames 11300...
870
+ [2023-02-23 10:05:39,842][11985] Num frames 11400...
871
+ [2023-02-23 10:05:39,950][11985] Num frames 11500...
872
+ [2023-02-23 10:05:40,040][11985] Avg episode rewards: #0: 33.483, true rewards: #0: 12.817
873
+ [2023-02-23 10:05:40,042][11985] Avg episode reward: 33.483, avg true_objective: 12.817
874
+ [2023-02-23 10:05:40,116][11985] Num frames 11600...
875
+ [2023-02-23 10:05:40,224][11985] Num frames 11700...
876
+ [2023-02-23 10:05:40,325][11985] Num frames 11800...
877
+ [2023-02-23 10:05:40,434][11985] Num frames 11900...
878
+ [2023-02-23 10:05:40,533][11985] Num frames 12000...
879
+ [2023-02-23 10:05:40,635][11985] Num frames 12100...
880
+ [2023-02-23 10:05:40,736][11985] Num frames 12200...
881
+ [2023-02-23 10:05:40,836][11985] Num frames 12300...
882
+ [2023-02-23 10:05:40,941][11985] Num frames 12400...
883
+ [2023-02-23 10:05:41,045][11985] Num frames 12500...
884
+ [2023-02-23 10:05:41,156][11985] Num frames 12600...
885
+ [2023-02-23 10:05:41,266][11985] Num frames 12700...
886
+ [2023-02-23 10:05:41,372][11985] Num frames 12800...
887
+ [2023-02-23 10:05:41,475][11985] Num frames 12900...
888
+ [2023-02-23 10:05:41,577][11985] Num frames 13000...
889
+ [2023-02-23 10:05:41,682][11985] Num frames 13100...
890
+ [2023-02-23 10:05:41,818][11985] Avg episode rewards: #0: 33.978, true rewards: #0: 13.178
891
+ [2023-02-23 10:05:41,820][11985] Avg episode reward: 33.978, avg true_objective: 13.178
892
+ [2023-02-23 10:06:25,487][11985] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
893
+ [2023-02-23 10:07:58,227][11985] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
894
+ [2023-02-23 10:07:58,228][11985] Overriding arg 'num_workers' with value 1 passed from command line
895
+ [2023-02-23 10:07:58,230][11985] Adding new argument 'no_render'=True that is not in the saved config file!
896
+ [2023-02-23 10:07:58,232][11985] Adding new argument 'save_video'=True that is not in the saved config file!
897
+ [2023-02-23 10:07:58,234][11985] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
898
+ [2023-02-23 10:07:58,236][11985] Adding new argument 'video_name'=None that is not in the saved config file!
899
+ [2023-02-23 10:07:58,237][11985] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
900
+ [2023-02-23 10:07:58,240][11985] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
901
+ [2023-02-23 10:07:58,241][11985] Adding new argument 'push_to_hub'=True that is not in the saved config file!
902
+ [2023-02-23 10:07:58,242][11985] Adding new argument 'hf_repository'='moodlep/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
903
+ [2023-02-23 10:07:58,245][11985] Adding new argument 'policy_index'=0 that is not in the saved config file!
904
+ [2023-02-23 10:07:58,246][11985] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
905
+ [2023-02-23 10:07:58,248][11985] Adding new argument 'train_script'=None that is not in the saved config file!
906
+ [2023-02-23 10:07:58,250][11985] Adding new argument 'enjoy_script'=None that is not in the saved config file!
907
+ [2023-02-23 10:07:58,251][11985] Using frameskip 1 and render_action_repeat=4 for evaluation
908
+ [2023-02-23 10:07:58,269][11985] RunningMeanStd input shape: (3, 72, 128)
909
+ [2023-02-23 10:07:58,271][11985] RunningMeanStd input shape: (1,)
910
+ [2023-02-23 10:07:58,284][11985] ConvEncoder: input_channels=3
911
+ [2023-02-23 10:07:58,318][11985] Conv encoder output size: 512
912
+ [2023-02-23 10:07:58,319][11985] Policy head output size: 512
913
+ [2023-02-23 10:07:58,340][11985] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
914
+ [2023-02-23 10:07:58,739][11985] Num frames 100...
915
+ [2023-02-23 10:07:58,839][11985] Num frames 200...
916
+ [2023-02-23 10:07:58,936][11985] Num frames 300...
917
+ [2023-02-23 10:07:59,007][11985] Avg episode rewards: #0: 6.160, true rewards: #0: 3.160
918
+ [2023-02-23 10:07:59,009][11985] Avg episode reward: 6.160, avg true_objective: 3.160
919
+ [2023-02-23 10:07:59,101][11985] Num frames 400...
920
+ [2023-02-23 10:07:59,205][11985] Num frames 500...
921
+ [2023-02-23 10:07:59,304][11985] Num frames 600...
922
+ [2023-02-23 10:07:59,413][11985] Num frames 700...
923
+ [2023-02-23 10:07:59,514][11985] Num frames 800...
924
+ [2023-02-23 10:07:59,626][11985] Num frames 900...
925
+ [2023-02-23 10:07:59,728][11985] Num frames 1000...
926
+ [2023-02-23 10:07:59,835][11985] Num frames 1100...
927
+ [2023-02-23 10:07:59,973][11985] Avg episode rewards: #0: 14.400, true rewards: #0: 5.900
928
+ [2023-02-23 10:07:59,975][11985] Avg episode reward: 14.400, avg true_objective: 5.900
929
+ [2023-02-23 10:08:00,002][11985] Num frames 1200...
930
+ [2023-02-23 10:08:00,108][11985] Num frames 1300...
931
+ [2023-02-23 10:08:00,217][11985] Num frames 1400...
932
+ [2023-02-23 10:08:00,318][11985] Num frames 1500...
933
+ [2023-02-23 10:08:00,421][11985] Num frames 1600...
934
+ [2023-02-23 10:08:00,530][11985] Num frames 1700...
935
+ [2023-02-23 10:08:00,636][11985] Num frames 1800...
936
+ [2023-02-23 10:08:00,738][11985] Num frames 1900...
937
+ [2023-02-23 10:08:00,825][11985] Avg episode rewards: #0: 14.770, true rewards: #0: 6.437
938
+ [2023-02-23 10:08:00,827][11985] Avg episode reward: 14.770, avg true_objective: 6.437
939
+ [2023-02-23 10:08:00,904][11985] Num frames 2000...
940
+ [2023-02-23 10:08:01,013][11985] Num frames 2100...
941
+ [2023-02-23 10:08:01,114][11985] Num frames 2200...
942
+ [2023-02-23 10:08:01,229][11985] Num frames 2300...
943
+ [2023-02-23 10:08:01,332][11985] Num frames 2400...
944
+ [2023-02-23 10:08:01,435][11985] Num frames 2500...
945
+ [2023-02-23 10:08:01,546][11985] Num frames 2600...
946
+ [2023-02-23 10:08:01,694][11985] Avg episode rewards: #0: 15.210, true rewards: #0: 6.710
947
+ [2023-02-23 10:08:01,696][11985] Avg episode reward: 15.210, avg true_objective: 6.710
948
+ [2023-02-23 10:08:01,716][11985] Num frames 2700...
949
+ [2023-02-23 10:08:01,822][11985] Num frames 2800...
950
+ [2023-02-23 10:08:01,943][11985] Num frames 2900...
951
+ [2023-02-23 10:08:02,063][11985] Num frames 3000...
952
+ [2023-02-23 10:08:02,176][11985] Num frames 3100...
953
+ [2023-02-23 10:08:02,278][11985] Num frames 3200...
954
+ [2023-02-23 10:08:02,379][11985] Num frames 3300...
955
+ [2023-02-23 10:08:02,482][11985] Num frames 3400...
956
+ [2023-02-23 10:08:02,592][11985] Num frames 3500...
957
+ [2023-02-23 10:08:02,698][11985] Num frames 3600...
958
+ [2023-02-23 10:08:02,800][11985] Num frames 3700...
959
+ [2023-02-23 10:08:02,902][11985] Num frames 3800...
960
+ [2023-02-23 10:08:03,008][11985] Num frames 3900...
961
+ [2023-02-23 10:08:03,114][11985] Num frames 4000...
962
+ [2023-02-23 10:08:03,222][11985] Num frames 4100...
963
+ [2023-02-23 10:08:03,334][11985] Num frames 4200...
964
+ [2023-02-23 10:08:03,450][11985] Num frames 4300...
965
+ [2023-02-23 10:08:03,508][11985] Avg episode rewards: #0: 20.202, true rewards: #0: 8.602
966
+ [2023-02-23 10:08:03,510][11985] Avg episode reward: 20.202, avg true_objective: 8.602
967
+ [2023-02-23 10:08:03,615][11985] Num frames 4400...
968
+ [2023-02-23 10:08:03,720][11985] Num frames 4500...
969
+ [2023-02-23 10:08:03,829][11985] Num frames 4600...
970
+ [2023-02-23 10:08:03,936][11985] Num frames 4700...
971
+ [2023-02-23 10:08:04,043][11985] Num frames 4800...
972
+ [2023-02-23 10:08:04,154][11985] Num frames 4900...
973
+ [2023-02-23 10:08:04,281][11985] Num frames 5000...
974
+ [2023-02-23 10:08:04,402][11985] Num frames 5100...
975
+ [2023-02-23 10:08:04,515][11985] Num frames 5200...
976
+ [2023-02-23 10:08:04,629][11985] Num frames 5300...
977
+ [2023-02-23 10:08:04,735][11985] Num frames 5400...
978
+ [2023-02-23 10:08:04,846][11985] Num frames 5500...
979
+ [2023-02-23 10:08:04,953][11985] Num frames 5600...
980
+ [2023-02-23 10:08:05,065][11985] Num frames 5700...
981
+ [2023-02-23 10:08:05,172][11985] Num frames 5800...
982
+ [2023-02-23 10:08:05,325][11985] Avg episode rewards: #0: 23.480, true rewards: #0: 9.813
983
+ [2023-02-23 10:08:05,327][11985] Avg episode reward: 23.480, avg true_objective: 9.813
984
+ [2023-02-23 10:08:05,342][11985] Num frames 5900...
985
+ [2023-02-23 10:08:05,447][11985] Num frames 6000...
986
+ [2023-02-23 10:08:05,547][11985] Num frames 6100...
987
+ [2023-02-23 10:08:05,651][11985] Num frames 6200...
988
+ [2023-02-23 10:08:05,751][11985] Num frames 6300...
989
+ [2023-02-23 10:08:05,852][11985] Num frames 6400...
990
+ [2023-02-23 10:08:05,958][11985] Num frames 6500...
991
+ [2023-02-23 10:08:06,065][11985] Num frames 6600...
992
+ [2023-02-23 10:08:06,170][11985] Num frames 6700...
993
+ [2023-02-23 10:08:06,275][11985] Num frames 6800...
994
+ [2023-02-23 10:08:06,402][11985] Num frames 6900...
995
+ [2023-02-23 10:08:06,518][11985] Num frames 7000...
996
+ [2023-02-23 10:08:06,635][11985] Num frames 7100...
997
+ [2023-02-23 10:08:06,751][11985] Num frames 7200...
998
+ [2023-02-23 10:08:06,866][11985] Num frames 7300...
999
+ [2023-02-23 10:08:06,984][11985] Num frames 7400...
1000
+ [2023-02-23 10:08:07,105][11985] Num frames 7500...
1001
+ [2023-02-23 10:08:07,221][11985] Avg episode rewards: #0: 26.217, true rewards: #0: 10.789
1002
+ [2023-02-23 10:08:07,223][11985] Avg episode reward: 26.217, avg true_objective: 10.789
1003
+ [2023-02-23 10:08:07,284][11985] Num frames 7600...
1004
+ [2023-02-23 10:08:07,394][11985] Num frames 7700...
1005
+ [2023-02-23 10:08:07,512][11985] Num frames 7800...
1006
+ [2023-02-23 10:08:07,642][11985] Num frames 7900...
1007
+ [2023-02-23 10:08:07,762][11985] Num frames 8000...
1008
+ [2023-02-23 10:08:07,877][11985] Num frames 8100...
1009
+ [2023-02-23 10:08:07,989][11985] Num frames 8200...
1010
+ [2023-02-23 10:08:08,103][11985] Num frames 8300...
1011
+ [2023-02-23 10:08:08,218][11985] Num frames 8400...
1012
+ [2023-02-23 10:08:08,333][11985] Num frames 8500...
1013
+ [2023-02-23 10:08:08,445][11985] Num frames 8600...
1014
+ [2023-02-23 10:08:08,554][11985] Num frames 8700...
1015
+ [2023-02-23 10:08:08,666][11985] Num frames 8800...
1016
+ [2023-02-23 10:08:08,775][11985] Num frames 8900...
1017
+ [2023-02-23 10:08:08,888][11985] Num frames 9000...
1018
+ [2023-02-23 10:08:08,989][11985] Num frames 9100...
1019
+ [2023-02-23 10:08:09,108][11985] Num frames 9200...
1020
+ [2023-02-23 10:08:09,220][11985] Num frames 9300...
1021
+ [2023-02-23 10:08:09,328][11985] Num frames 9400...
1022
+ [2023-02-23 10:08:09,426][11985] Num frames 9500...
1023
+ [2023-02-23 10:08:09,537][11985] Num frames 9600...
1024
+ [2023-02-23 10:08:09,648][11985] Avg episode rewards: #0: 29.565, true rewards: #0: 12.065
1025
+ [2023-02-23 10:08:09,650][11985] Avg episode reward: 29.565, avg true_objective: 12.065
1026
+ [2023-02-23 10:08:09,703][11985] Num frames 9700...
1027
+ [2023-02-23 10:08:09,813][11985] Num frames 9800...
1028
+ [2023-02-23 10:08:09,922][11985] Num frames 9900...
1029
+ [2023-02-23 10:08:10,030][11985] Num frames 10000...
1030
+ [2023-02-23 10:08:10,133][11985] Num frames 10100...
1031
+ [2023-02-23 10:08:10,245][11985] Num frames 10200...
1032
+ [2023-02-23 10:08:10,356][11985] Num frames 10300...
1033
+ [2023-02-23 10:08:10,415][11985] Avg episode rewards: #0: 28.447, true rewards: #0: 11.447
1034
+ [2023-02-23 10:08:10,417][11985] Avg episode reward: 28.447, avg true_objective: 11.447
1035
+ [2023-02-23 10:08:10,517][11985] Num frames 10400...
1036
+ [2023-02-23 10:08:10,622][11985] Num frames 10500...
1037
+ [2023-02-23 10:08:10,733][11985] Num frames 10600...
1038
+ [2023-02-23 10:08:10,835][11985] Num frames 10700...
1039
+ [2023-02-23 10:08:10,941][11985] Num frames 10800...
1040
+ [2023-02-23 10:08:11,043][11985] Num frames 10900...
1041
+ [2023-02-23 10:08:11,147][11985] Num frames 11000...
1042
+ [2023-02-23 10:08:11,247][11985] Num frames 11100...
1043
+ [2023-02-23 10:08:11,358][11985] Num frames 11200...
1044
+ [2023-02-23 10:08:11,466][11985] Num frames 11300...
1045
+ [2023-02-23 10:08:11,574][11985] Num frames 11400...
1046
+ [2023-02-23 10:08:11,690][11985] Num frames 11500...
1047
+ [2023-02-23 10:08:11,796][11985] Num frames 11600...
1048
+ [2023-02-23 10:08:11,907][11985] Num frames 11700...
1049
+ [2023-02-23 10:08:12,012][11985] Num frames 11800...
1050
+ [2023-02-23 10:08:12,116][11985] Num frames 11900...
1051
+ [2023-02-23 10:08:12,219][11985] Num frames 12000...
1052
+ [2023-02-23 10:08:12,330][11985] Num frames 12100...
1053
+ [2023-02-23 10:08:12,413][11985] Avg episode rewards: #0: 30.325, true rewards: #0: 12.125
1054
+ [2023-02-23 10:08:12,415][11985] Avg episode reward: 30.325, avg true_objective: 12.125
1055
+ [2023-02-23 10:08:52,893][11985] Replay video saved to /content/train_dir/default_experiment/replay.mp4!