Hugging Face's logo

amanoyaku
/
rl_course_vizdoom_health_gathering_supreme

Reinforcement Learning
sample-factory
TensorBoard
deep-reinforcement-learning
Eval Results
Model card Files Metrics Community
amanoyaku commited on
Commit
3fbc60f
·
verified ·
1 Parent(s): e2e7094

Upload folder using huggingface_hub

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. .summary/0/events.out.tfevents.1735539931.86790b443086 +3 -0
  3. README.md +56 -0
  4. checkpoint_p0/best_000000961_3936256_reward_24.588.pth +3 -0
  5. checkpoint_p0/checkpoint_000000493_2019328.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 +720 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ replay.mp4 filter=lfs diff=lfs merge=lfs -text
.summary/0/events.out.tfevents.1735539931.86790b443086 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00defbd4f92998dd448d52a8b492243ff9b3b8e1f1173d2ce6b5fe2ee6e6bc41
3
+ size 199798
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: 9.37 +/- 4.81
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 amanoyaku/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_000000961_3936256_reward_24.588.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b23be1b9fc584345742c9b26af2702dcf4a77d340ef267151fda97a674852183
3
+ size 34929051
checkpoint_p0/checkpoint_000000493_2019328.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb0e1c4b2cff3526394cdfa8e6f0788b7669894f195eef69da9bbbeea9dca4c6
3
+ size 34929541
checkpoint_p0/checkpoint_000000978_4005888.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32c23236e4f4968c9d04dd2e1be29c2ede946dba325926ab496cff81fed9669e
3
+ size 34929541
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 sha256:45cd84a6ee1d3017086bda38ef0acb5ecfccbdeef015939c27707b8b1652b347
3
+ size 17223963
sf_log.txt ADDED
@@ -0,0 +1,720 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2024-12-30 06:25:36,744][00704] Saving configuration to /content/train_dir/default_experiment/config.json...
2
+ [2024-12-30 06:25:36,747][00704] Rollout worker 0 uses device cpu
3
+ [2024-12-30 06:25:36,748][00704] Rollout worker 1 uses device cpu
4
+ [2024-12-30 06:25:36,749][00704] Rollout worker 2 uses device cpu
5
+ [2024-12-30 06:25:36,750][00704] Rollout worker 3 uses device cpu
6
+ [2024-12-30 06:25:36,752][00704] Rollout worker 4 uses device cpu
7
+ [2024-12-30 06:25:36,753][00704] Rollout worker 5 uses device cpu
8
+ [2024-12-30 06:25:36,754][00704] Rollout worker 6 uses device cpu
9
+ [2024-12-30 06:25:36,755][00704] Rollout worker 7 uses device cpu
10
+ [2024-12-30 06:25:36,860][00704] Using GPUs [0] for process 0 (actually maps to GPUs [0])
11
+ [2024-12-30 06:25:36,862][00704] InferenceWorker_p0-w0: min num requests: 2
12
+ [2024-12-30 06:25:36,894][00704] Starting all processes...
13
+ [2024-12-30 06:25:36,896][00704] Starting process learner_proc0
14
+ [2024-12-30 06:25:36,941][00704] Starting all processes...
15
+ [2024-12-30 06:25:36,947][00704] Starting process inference_proc0-0
16
+ [2024-12-30 06:25:36,948][00704] Starting process rollout_proc0
17
+ [2024-12-30 06:25:36,951][00704] Starting process rollout_proc1
18
+ [2024-12-30 06:25:36,951][00704] Starting process rollout_proc2
19
+ [2024-12-30 06:25:36,954][00704] Starting process rollout_proc3
20
+ [2024-12-30 06:25:36,959][00704] Starting process rollout_proc4
21
+ [2024-12-30 06:25:36,965][00704] Starting process rollout_proc5
22
+ [2024-12-30 06:25:36,968][00704] Starting process rollout_proc6
23
+ [2024-12-30 06:25:36,970][00704] Starting process rollout_proc7
24
+ [2024-12-30 06:25:39,825][02856] Worker 2 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
25
+ [2024-12-30 06:25:39,874][02858] Worker 4 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
26
+ [2024-12-30 06:25:39,880][02853] Using GPUs [0] for process 0 (actually maps to GPUs [0])
27
+ [2024-12-30 06:25:39,881][02853] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
28
+ [2024-12-30 06:25:39,900][02853] Num visible devices: 1
29
+ [2024-12-30 06:25:40,089][02854] Worker 0 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
30
+ [2024-12-30 06:25:40,150][02840] Using GPUs [0] for process 0 (actually maps to GPUs [0])
31
+ [2024-12-30 06:25:40,150][02840] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
32
+ [2024-12-30 06:25:40,165][02840] Num visible devices: 1
33
+ [2024-12-30 06:25:40,188][02840] Starting seed is not provided
34
+ [2024-12-30 06:25:40,189][02840] Using GPUs [0] for process 0 (actually maps to GPUs [0])
35
+ [2024-12-30 06:25:40,189][02840] Initializing actor-critic model on device cuda:0
36
+ [2024-12-30 06:25:40,189][02840] RunningMeanStd input shape: (3, 72, 128)
37
+ [2024-12-30 06:25:40,192][02840] RunningMeanStd input shape: (1,)
38
+ [2024-12-30 06:25:40,205][02840] ConvEncoder: input_channels=3
39
+ [2024-12-30 06:25:40,214][02857] Worker 3 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
40
+ [2024-12-30 06:25:40,394][02855] Worker 1 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
41
+ [2024-12-30 06:25:40,439][02859] Worker 5 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
42
+ [2024-12-30 06:25:40,471][02860] Worker 6 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
43
+ [2024-12-30 06:25:40,510][02861] Worker 7 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
44
+ [2024-12-30 06:25:40,510][02840] Conv encoder output size: 512
45
+ [2024-12-30 06:25:40,511][02840] Policy head output size: 512
46
+ [2024-12-30 06:25:40,563][02840] Created Actor Critic model with architecture:
47
+ [2024-12-30 06:25:40,564][02840] ActorCriticSharedWeights(
48
+ (obs_normalizer): ObservationNormalizer(
49
+ (running_mean_std): RunningMeanStdDictInPlace(
50
+ (running_mean_std): ModuleDict(
51
+ (obs): RunningMeanStdInPlace()
52
+ )
53
+ )
54
+ )
55
+ (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
56
+ (encoder): VizdoomEncoder(
57
+ (basic_encoder): ConvEncoder(
58
+ (enc): RecursiveScriptModule(
59
+ original_name=ConvEncoderImpl
60
+ (conv_head): RecursiveScriptModule(
61
+ original_name=Sequential
62
+ (0): RecursiveScriptModule(original_name=Conv2d)
63
+ (1): RecursiveScriptModule(original_name=ELU)
64
+ (2): RecursiveScriptModule(original_name=Conv2d)
65
+ (3): RecursiveScriptModule(original_name=ELU)
66
+ (4): RecursiveScriptModule(original_name=Conv2d)
67
+ (5): RecursiveScriptModule(original_name=ELU)
68
+ )
69
+ (mlp_layers): RecursiveScriptModule(
70
+ original_name=Sequential
71
+ (0): RecursiveScriptModule(original_name=Linear)
72
+ (1): RecursiveScriptModule(original_name=ELU)
73
+ )
74
+ )
75
+ )
76
+ )
77
+ (core): ModelCoreRNN(
78
+ (core): GRU(512, 512)
79
+ )
80
+ (decoder): MlpDecoder(
81
+ (mlp): Identity()
82
+ )
83
+ (critic_linear): Linear(in_features=512, out_features=1, bias=True)
84
+ (action_parameterization): ActionParameterizationDefault(
85
+ (distribution_linear): Linear(in_features=512, out_features=5, bias=True)
86
+ )
87
+ )
88
+ [2024-12-30 06:25:40,838][02840] Using optimizer <class 'torch.optim.adam.Adam'>
89
+ [2024-12-30 06:25:44,231][02840] No checkpoints found
90
+ [2024-12-30 06:25:44,231][02840] Did not load from checkpoint, starting from scratch!
91
+ [2024-12-30 06:25:44,232][02840] Initialized policy 0 weights for model version 0
92
+ [2024-12-30 06:25:44,235][02840] LearnerWorker_p0 finished initialization!
93
+ [2024-12-30 06:25:44,235][02840] Using GPUs [0] for process 0 (actually maps to GPUs [0])
94
+ [2024-12-30 06:25:44,309][02853] RunningMeanStd input shape: (3, 72, 128)
95
+ [2024-12-30 06:25:44,310][02853] RunningMeanStd input shape: (1,)
96
+ [2024-12-30 06:25:44,323][02853] ConvEncoder: input_channels=3
97
+ [2024-12-30 06:25:44,430][02853] Conv encoder output size: 512
98
+ [2024-12-30 06:25:44,431][02853] Policy head output size: 512
99
+ [2024-12-30 06:25:44,482][00704] Inference worker 0-0 is ready!
100
+ [2024-12-30 06:25:44,483][00704] All inference workers are ready! Signal rollout workers to start!
101
+ [2024-12-30 06:25:44,517][02861] Doom resolution: 160x120, resize resolution: (128, 72)
102
+ [2024-12-30 06:25:44,517][02858] Doom resolution: 160x120, resize resolution: (128, 72)
103
+ [2024-12-30 06:25:44,517][02857] Doom resolution: 160x120, resize resolution: (128, 72)
104
+ [2024-12-30 06:25:44,517][02860] Doom resolution: 160x120, resize resolution: (128, 72)
105
+ [2024-12-30 06:25:44,537][02855] Doom resolution: 160x120, resize resolution: (128, 72)
106
+ [2024-12-30 06:25:44,537][02854] Doom resolution: 160x120, resize resolution: (128, 72)
107
+ [2024-12-30 06:25:44,537][02859] Doom resolution: 160x120, resize resolution: (128, 72)
108
+ [2024-12-30 06:25:44,537][02856] Doom resolution: 160x120, resize resolution: (128, 72)
109
+ [2024-12-30 06:25:44,831][02856] Decorrelating experience for 0 frames...
110
+ [2024-12-30 06:25:44,831][02859] Decorrelating experience for 0 frames...
111
+ [2024-12-30 06:25:44,831][02858] Decorrelating experience for 0 frames...
112
+ [2024-12-30 06:25:44,831][02861] Decorrelating experience for 0 frames...
113
+ [2024-12-30 06:25:44,831][02860] Decorrelating experience for 0 frames...
114
+ [2024-12-30 06:25:44,906][02857] Decorrelating experience for 0 frames...
115
+ [2024-12-30 06:25:44,941][02855] Decorrelating experience for 0 frames...
116
+ [2024-12-30 06:25:45,070][02859] Decorrelating experience for 32 frames...
117
+ [2024-12-30 06:25:45,071][02856] Decorrelating experience for 32 frames...
118
+ [2024-12-30 06:25:45,095][02854] Decorrelating experience for 0 frames...
119
+ [2024-12-30 06:25:45,105][02861] Decorrelating experience for 32 frames...
120
+ [2024-12-30 06:25:45,186][02857] Decorrelating experience for 32 frames...
121
+ [2024-12-30 06:25:45,329][02860] Decorrelating experience for 32 frames...
122
+ [2024-12-30 06:25:45,338][02854] Decorrelating experience for 32 frames...
123
+ [2024-12-30 06:25:45,376][02858] Decorrelating experience for 32 frames...
124
+ [2024-12-30 06:25:45,383][02855] Decorrelating experience for 32 frames...
125
+ [2024-12-30 06:25:45,458][02856] Decorrelating experience for 64 frames...
126
+ [2024-12-30 06:25:45,475][02859] Decorrelating experience for 64 frames...
127
+ [2024-12-30 06:25:45,634][02857] Decorrelating experience for 64 frames...
128
+ [2024-12-30 06:25:45,695][02860] Decorrelating experience for 64 frames...
129
+ [2024-12-30 06:25:45,696][02854] Decorrelating experience for 64 frames...
130
+ [2024-12-30 06:25:45,699][02858] Decorrelating experience for 64 frames...
131
+ [2024-12-30 06:25:45,770][02856] Decorrelating experience for 96 frames...
132
+ [2024-12-30 06:25:45,919][02855] Decorrelating experience for 64 frames...
133
+ [2024-12-30 06:25:45,968][02857] Decorrelating experience for 96 frames...
134
+ [2024-12-30 06:25:45,990][02854] Decorrelating experience for 96 frames...
135
+ [2024-12-30 06:25:46,004][02858] Decorrelating experience for 96 frames...
136
+ [2024-12-30 06:25:46,017][02859] Decorrelating experience for 96 frames...
137
+ [2024-12-30 06:25:46,065][02860] Decorrelating experience for 96 frames...
138
+ [2024-12-30 06:25:46,214][02861] Decorrelating experience for 64 frames...
139
+ [2024-12-30 06:25:46,242][02855] Decorrelating experience for 96 frames...
140
+ [2024-12-30 06:25:46,487][02861] Decorrelating experience for 96 frames...
141
+ [2024-12-30 06:25:46,690][00704] 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)
142
+ [2024-12-30 06:25:48,011][02840] Signal inference workers to stop experience collection...
143
+ [2024-12-30 06:25:48,016][02853] InferenceWorker_p0-w0: stopping experience collection
144
+ [2024-12-30 06:25:50,427][02840] Signal inference workers to resume experience collection...
145
+ [2024-12-30 06:25:50,428][02853] InferenceWorker_p0-w0: resuming experience collection
146
+ [2024-12-30 06:25:51,690][00704] Fps is (10 sec: 4915.1, 60 sec: 4915.1, 300 sec: 4915.1). Total num frames: 24576. Throughput: 0: 512.4. Samples: 2562. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
147
+ [2024-12-30 06:25:51,692][00704] Avg episode reward: [(0, '3.529')]
148
+ [2024-12-30 06:25:52,337][02853] Updated weights for policy 0, policy_version 10 (0.0152)
149
+ [2024-12-30 06:25:54,480][02853] Updated weights for policy 0, policy_version 20 (0.0013)
150
+ [2024-12-30 06:25:56,547][02853] Updated weights for policy 0, policy_version 30 (0.0014)
151
+ [2024-12-30 06:25:56,690][00704] Fps is (10 sec: 12287.9, 60 sec: 12287.9, 300 sec: 12287.9). Total num frames: 122880. Throughput: 0: 3076.4. Samples: 30764. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
152
+ [2024-12-30 06:25:56,693][00704] Avg episode reward: [(0, '4.386')]
153
+ [2024-12-30 06:25:56,695][02840] Saving new best policy, reward=4.386!
154
+ [2024-12-30 06:25:56,852][00704] Heartbeat connected on Batcher_0
155
+ [2024-12-30 06:25:56,855][00704] Heartbeat connected on LearnerWorker_p0
156
+ [2024-12-30 06:25:56,865][00704] Heartbeat connected on InferenceWorker_p0-w0
157
+ [2024-12-30 06:25:56,868][00704] Heartbeat connected on RolloutWorker_w0
158
+ [2024-12-30 06:25:56,874][00704] Heartbeat connected on RolloutWorker_w1
159
+ [2024-12-30 06:25:56,877][00704] Heartbeat connected on RolloutWorker_w2
160
+ [2024-12-30 06:25:56,883][00704] Heartbeat connected on RolloutWorker_w4
161
+ [2024-12-30 06:25:56,884][00704] Heartbeat connected on RolloutWorker_w3
162
+ [2024-12-30 06:25:56,887][00704] Heartbeat connected on RolloutWorker_w5
163
+ [2024-12-30 06:25:56,890][00704] Heartbeat connected on RolloutWorker_w6
164
+ [2024-12-30 06:25:56,893][00704] Heartbeat connected on RolloutWorker_w7
165
+ [2024-12-30 06:25:58,595][02853] Updated weights for policy 0, policy_version 40 (0.0013)
166
+ [2024-12-30 06:26:00,626][02853] Updated weights for policy 0, policy_version 50 (0.0013)
167
+ [2024-12-30 06:26:01,690][00704] Fps is (10 sec: 20069.8, 60 sec: 15018.3, 300 sec: 15018.3). Total num frames: 225280. Throughput: 0: 3052.6. Samples: 45790. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
168
+ [2024-12-30 06:26:01,693][00704] Avg episode reward: [(0, '4.307')]
169
+ [2024-12-30 06:26:02,662][02853] Updated weights for policy 0, policy_version 60 (0.0013)
170
+ [2024-12-30 06:26:04,704][02853] Updated weights for policy 0, policy_version 70 (0.0013)
171
+ [2024-12-30 06:26:06,690][00704] Fps is (10 sec: 20070.6, 60 sec: 16179.2, 300 sec: 16179.2). Total num frames: 323584. Throughput: 0: 3806.3. Samples: 76126. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
172
+ [2024-12-30 06:26:06,692][00704] Avg episode reward: [(0, '4.323')]
173
+ [2024-12-30 06:26:06,757][02853] Updated weights for policy 0, policy_version 80 (0.0013)
174
+ [2024-12-30 06:26:08,888][02853] Updated weights for policy 0, policy_version 90 (0.0014)
175
+ [2024-12-30 06:26:10,953][02853] Updated weights for policy 0, policy_version 100 (0.0013)
176
+ [2024-12-30 06:26:11,690][00704] Fps is (10 sec: 19661.3, 60 sec: 16875.5, 300 sec: 16875.5). Total num frames: 421888. Throughput: 0: 4215.8. Samples: 105396. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
177
+ [2024-12-30 06:26:11,692][00704] Avg episode reward: [(0, '4.506')]
178
+ [2024-12-30 06:26:11,701][02840] Saving new best policy, reward=4.506!
179
+ [2024-12-30 06:26:13,048][02853] Updated weights for policy 0, policy_version 110 (0.0013)
180
+ [2024-12-30 06:26:15,048][02853] Updated weights for policy 0, policy_version 120 (0.0013)
181
+ [2024-12-30 06:26:16,690][00704] Fps is (10 sec: 20070.2, 60 sec: 17476.2, 300 sec: 17476.2). Total num frames: 524288. Throughput: 0: 4017.5. Samples: 120526. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
182
+ [2024-12-30 06:26:16,692][00704] Avg episode reward: [(0, '4.687')]
183
+ [2024-12-30 06:26:16,694][02840] Saving new best policy, reward=4.687!
184
+ [2024-12-30 06:26:17,064][02853] Updated weights for policy 0, policy_version 130 (0.0012)
185
+ [2024-12-30 06:26:19,090][02853] Updated weights for policy 0, policy_version 140 (0.0013)
186
+ [2024-12-30 06:26:21,173][02853] Updated weights for policy 0, policy_version 150 (0.0013)
187
+ [2024-12-30 06:26:21,690][00704] Fps is (10 sec: 20070.4, 60 sec: 17788.3, 300 sec: 17788.3). Total num frames: 622592. Throughput: 0: 4306.9. Samples: 150742. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
188
+ [2024-12-30 06:26:21,693][00704] Avg episode reward: [(0, '4.528')]
189
+ [2024-12-30 06:26:23,327][02853] Updated weights for policy 0, policy_version 160 (0.0013)
190
+ [2024-12-30 06:26:25,405][02853] Updated weights for policy 0, policy_version 170 (0.0013)
191
+ [2024-12-30 06:26:26,690][00704] Fps is (10 sec: 19660.8, 60 sec: 18022.4, 300 sec: 18022.4). Total num frames: 720896. Throughput: 0: 4506.1. Samples: 180246. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
192
+ [2024-12-30 06:26:26,693][00704] Avg episode reward: [(0, '4.670')]
193
+ [2024-12-30 06:26:27,401][02853] Updated weights for policy 0, policy_version 180 (0.0013)
194
+ [2024-12-30 06:26:29,410][02853] Updated weights for policy 0, policy_version 190 (0.0013)
195
+ [2024-12-30 06:26:31,440][02853] Updated weights for policy 0, policy_version 200 (0.0013)
196
+ [2024-12-30 06:26:31,690][00704] Fps is (10 sec: 20070.5, 60 sec: 18295.4, 300 sec: 18295.4). Total num frames: 823296. Throughput: 0: 4344.4. Samples: 195500. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
197
+ [2024-12-30 06:26:31,692][00704] Avg episode reward: [(0, '4.595')]
198
+ [2024-12-30 06:26:33,471][02853] Updated weights for policy 0, policy_version 210 (0.0013)
199
+ [2024-12-30 06:26:35,551][02853] Updated weights for policy 0, policy_version 220 (0.0013)
200
+ [2024-12-30 06:26:36,690][00704] Fps is (10 sec: 20070.2, 60 sec: 18431.9, 300 sec: 18431.9). Total num frames: 921600. Throughput: 0: 4956.1. Samples: 225588. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
201
+ [2024-12-30 06:26:36,692][00704] Avg episode reward: [(0, '4.693')]
202
+ [2024-12-30 06:26:36,694][02840] Saving new best policy, reward=4.693!
203
+ [2024-12-30 06:26:37,661][02853] Updated weights for policy 0, policy_version 230 (0.0013)
204
+ [2024-12-30 06:26:39,720][02853] Updated weights for policy 0, policy_version 240 (0.0013)
205
+ [2024-12-30 06:26:41,690][00704] Fps is (10 sec: 19660.8, 60 sec: 18543.7, 300 sec: 18543.7). Total num frames: 1019904. Throughput: 0: 4987.6. Samples: 255204. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
206
+ [2024-12-30 06:26:41,692][00704] Avg episode reward: [(0, '4.915')]
207
+ [2024-12-30 06:26:41,699][02840] Saving new best policy, reward=4.915!
208
+ [2024-12-30 06:26:41,815][02853] Updated weights for policy 0, policy_version 250 (0.0013)
209
+ [2024-12-30 06:26:43,832][02853] Updated weights for policy 0, policy_version 260 (0.0013)
210
+ [2024-12-30 06:26:45,874][02853] Updated weights for policy 0, policy_version 270 (0.0013)
211
+ [2024-12-30 06:26:46,690][00704] Fps is (10 sec: 20070.7, 60 sec: 18705.1, 300 sec: 18705.1). Total num frames: 1122304. Throughput: 0: 4984.8. Samples: 270102. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
212
+ [2024-12-30 06:26:46,692][00704] Avg episode reward: [(0, '5.186')]
213
+ [2024-12-30 06:26:46,694][02840] Saving new best policy, reward=5.186!
214
+ [2024-12-30 06:26:47,930][02853] Updated weights for policy 0, policy_version 280 (0.0013)
215
+ [2024-12-30 06:26:50,062][02853] Updated weights for policy 0, policy_version 290 (0.0012)
216
+ [2024-12-30 06:26:51,690][00704] Fps is (10 sec: 19660.9, 60 sec: 19865.6, 300 sec: 18715.6). Total num frames: 1216512. Throughput: 0: 4966.0. Samples: 299596. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
217
+ [2024-12-30 06:26:51,692][00704] Avg episode reward: [(0, '5.257')]
218
+ [2024-12-30 06:26:51,698][02840] Saving new best policy, reward=5.257!
219
+ [2024-12-30 06:26:52,181][02853] Updated weights for policy 0, policy_version 300 (0.0013)
220
+ [2024-12-30 06:26:54,202][02853] Updated weights for policy 0, policy_version 310 (0.0013)
221
+ [2024-12-30 06:26:56,240][02853] Updated weights for policy 0, policy_version 320 (0.0013)
222
+ [2024-12-30 06:26:56,690][00704] Fps is (10 sec: 19660.6, 60 sec: 19933.9, 300 sec: 18841.6). Total num frames: 1318912. Throughput: 0: 4982.5. Samples: 329608. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
223
+ [2024-12-30 06:26:56,692][00704] Avg episode reward: [(0, '5.775')]
224
+ [2024-12-30 06:26:56,695][02840] Saving new best policy, reward=5.775!
225
+ [2024-12-30 06:26:58,259][02853] Updated weights for policy 0, policy_version 330 (0.0013)
226
+ [2024-12-30 06:27:00,256][02853] Updated weights for policy 0, policy_version 340 (0.0013)
227
+ [2024-12-30 06:27:01,690][00704] Fps is (10 sec: 20480.1, 60 sec: 19934.0, 300 sec: 18950.8). Total num frames: 1421312. Throughput: 0: 4985.4. Samples: 344870. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
228
+ [2024-12-30 06:27:01,692][00704] Avg episode reward: [(0, '6.335')]
229
+ [2024-12-30 06:27:01,698][02840] Saving new best policy, reward=6.335!
230
+ [2024-12-30 06:27:02,292][02853] Updated weights for policy 0, policy_version 350 (0.0013)
231
+ [2024-12-30 06:27:04,442][02853] Updated weights for policy 0, policy_version 360 (0.0013)
232
+ [2024-12-30 06:27:06,536][02853] Updated weights for policy 0, policy_version 370 (0.0013)
233
+ [2024-12-30 06:27:06,690][00704] Fps is (10 sec: 19660.8, 60 sec: 19865.6, 300 sec: 18944.0). Total num frames: 1515520. Throughput: 0: 4970.9. Samples: 374434. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
234
+ [2024-12-30 06:27:06,692][00704] Avg episode reward: [(0, '6.616')]
235
+ [2024-12-30 06:27:06,695][02840] Saving new best policy, reward=6.616!
236
+ [2024-12-30 06:27:08,547][02853] Updated weights for policy 0, policy_version 380 (0.0012)
237
+ [2024-12-30 06:27:10,551][02853] Updated weights for policy 0, policy_version 390 (0.0013)
238
+ [2024-12-30 06:27:11,690][00704] Fps is (10 sec: 19660.9, 60 sec: 19933.9, 300 sec: 19034.4). Total num frames: 1617920. Throughput: 0: 4991.2. Samples: 404850. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
239
+ [2024-12-30 06:27:11,692][00704] Avg episode reward: [(0, '6.960')]
240
+ [2024-12-30 06:27:11,699][02840] Saving new best policy, reward=6.960!
241
+ [2024-12-30 06:27:12,587][02853] Updated weights for policy 0, policy_version 400 (0.0013)
242
+ [2024-12-30 06:27:14,615][02853] Updated weights for policy 0, policy_version 410 (0.0013)
243
+ [2024-12-30 06:27:16,634][02853] Updated weights for policy 0, policy_version 420 (0.0012)
244
+ [2024-12-30 06:27:16,690][00704] Fps is (10 sec: 20480.1, 60 sec: 19933.9, 300 sec: 19114.7). Total num frames: 1720320. Throughput: 0: 4988.4. Samples: 419980. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
245
+ [2024-12-30 06:27:16,692][00704] Avg episode reward: [(0, '7.441')]
246
+ [2024-12-30 06:27:16,694][02840] Saving new best policy, reward=7.441!
247
+ [2024-12-30 06:27:18,765][02853] Updated weights for policy 0, policy_version 430 (0.0013)
248
+ [2024-12-30 06:27:20,848][02853] Updated weights for policy 0, policy_version 440 (0.0013)
249
+ [2024-12-30 06:27:21,690][00704] Fps is (10 sec: 20070.2, 60 sec: 19933.9, 300 sec: 19143.4). Total num frames: 1818624. Throughput: 0: 4978.9. Samples: 449636. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
250
+ [2024-12-30 06:27:21,692][00704] Avg episode reward: [(0, '7.938')]
251
+ [2024-12-30 06:27:21,699][02840] Saving new best policy, reward=7.938!
252
+ [2024-12-30 06:27:22,842][02853] Updated weights for policy 0, policy_version 450 (0.0013)
253
+ [2024-12-30 06:27:24,862][02853] Updated weights for policy 0, policy_version 460 (0.0013)
254
+ [2024-12-30 06:27:26,690][00704] Fps is (10 sec: 20070.3, 60 sec: 20002.1, 300 sec: 19210.2). Total num frames: 1921024. Throughput: 0: 4996.8. Samples: 480058. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
255
+ [2024-12-30 06:27:26,692][00704] Avg episode reward: [(0, '8.467')]
256
+ [2024-12-30 06:27:26,695][02840] Saving new best policy, reward=8.467!
257
+ [2024-12-30 06:27:26,886][02853] Updated weights for policy 0, policy_version 470 (0.0013)
258
+ [2024-12-30 06:27:28,907][02853] Updated weights for policy 0, policy_version 480 (0.0013)
259
+ [2024-12-30 06:27:30,963][02853] Updated weights for policy 0, policy_version 490 (0.0013)
260
+ [2024-12-30 06:27:31,690][00704] Fps is (10 sec: 20070.2, 60 sec: 19933.8, 300 sec: 19231.7). Total num frames: 2019328. Throughput: 0: 5001.1. Samples: 495150. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
261
+ [2024-12-30 06:27:31,693][00704] Avg episode reward: [(0, '8.891')]
262
+ [2024-12-30 06:27:31,700][02840] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000493_2019328.pth...
263
+ [2024-12-30 06:27:31,772][02840] Saving new best policy, reward=8.891!
264
+ [2024-12-30 06:27:33,106][02853] Updated weights for policy 0, policy_version 500 (0.0014)
265
+ [2024-12-30 06:27:35,135][02853] Updated weights for policy 0, policy_version 510 (0.0013)
266
+ [2024-12-30 06:27:36,690][00704] Fps is (10 sec: 19660.7, 60 sec: 19933.9, 300 sec: 19251.2). Total num frames: 2117632. Throughput: 0: 5002.2. Samples: 524694. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
267
+ [2024-12-30 06:27:36,692][00704] Avg episode reward: [(0, '10.369')]
268
+ [2024-12-30 06:27:36,694][02840] Saving new best policy, reward=10.369!
269
+ [2024-12-30 06:27:37,152][02853] Updated weights for policy 0, policy_version 520 (0.0013)
270
+ [2024-12-30 06:27:39,157][02853] Updated weights for policy 0, policy_version 530 (0.0013)
271
+ [2024-12-30 06:27:41,181][02853] Updated weights for policy 0, policy_version 540 (0.0013)
272
+ [2024-12-30 06:27:41,690][00704] Fps is (10 sec: 20070.5, 60 sec: 20002.1, 300 sec: 19304.6). Total num frames: 2220032. Throughput: 0: 5013.8. Samples: 555228. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
273
+ [2024-12-30 06:27:41,693][00704] Avg episode reward: [(0, '12.016')]
274
+ [2024-12-30 06:27:41,699][02840] Saving new best policy, reward=12.016!
275
+ [2024-12-30 06:27:43,206][02853] Updated weights for policy 0, policy_version 550 (0.0013)
276
+ [2024-12-30 06:27:45,233][02853] Updated weights for policy 0, policy_version 560 (0.0013)
277
+ [2024-12-30 06:27:46,690][00704] Fps is (10 sec: 20480.3, 60 sec: 20002.1, 300 sec: 19353.6). Total num frames: 2322432. Throughput: 0: 5010.7. Samples: 570352. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
278
+ [2024-12-30 06:27:46,692][00704] Avg episode reward: [(0, '11.994')]
279
+ [2024-12-30 06:27:47,307][02853] Updated weights for policy 0, policy_version 570 (0.0013)
280
+ [2024-12-30 06:27:49,359][02853] Updated weights for policy 0, policy_version 580 (0.0013)
281
+ [2024-12-30 06:27:51,383][02853] Updated weights for policy 0, policy_version 590 (0.0013)
282
+ [2024-12-30 06:27:51,690][00704] Fps is (10 sec: 20070.3, 60 sec: 20070.4, 300 sec: 19365.9). Total num frames: 2420736. Throughput: 0: 5020.0. Samples: 600334. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
283
+ [2024-12-30 06:27:51,693][00704] Avg episode reward: [(0, '13.268')]
284
+ [2024-12-30 06:27:51,699][02840] Saving new best policy, reward=13.268!
285
+ [2024-12-30 06:27:53,398][02853] Updated weights for policy 0, policy_version 600 (0.0013)
286
+ [2024-12-30 06:27:55,394][02853] Updated weights for policy 0, policy_version 610 (0.0013)
287
+ [2024-12-30 06:27:56,690][00704] Fps is (10 sec: 20070.3, 60 sec: 20070.4, 300 sec: 19408.7). Total num frames: 2523136. Throughput: 0: 5023.7. Samples: 630918. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
288
+ [2024-12-30 06:27:56,692][00704] Avg episode reward: [(0, '15.725')]
289
+ [2024-12-30 06:27:56,694][02840] Saving new best policy, reward=15.725!
290
+ [2024-12-30 06:27:57,394][02853] Updated weights for policy 0, policy_version 620 (0.0013)
291
+ [2024-12-30 06:27:59,504][02853] Updated weights for policy 0, policy_version 630 (0.0014)
292
+ [2024-12-30 06:28:01,611][02853] Updated weights for policy 0, policy_version 640 (0.0013)
293
+ [2024-12-30 06:28:01,690][00704] Fps is (10 sec: 20070.4, 60 sec: 20002.1, 300 sec: 19418.1). Total num frames: 2621440. Throughput: 0: 5018.4. Samples: 645810. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
294
+ [2024-12-30 06:28:01,693][00704] Avg episode reward: [(0, '14.407')]
295
+ [2024-12-30 06:28:03,620][02853] Updated weights for policy 0, policy_version 650 (0.0013)
296
+ [2024-12-30 06:28:05,636][02853] Updated weights for policy 0, policy_version 660 (0.0012)
297
+ [2024-12-30 06:28:06,690][00704] Fps is (10 sec: 20070.6, 60 sec: 20138.7, 300 sec: 19456.0). Total num frames: 2723840. Throughput: 0: 5027.9. Samples: 675890. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
298
+ [2024-12-30 06:28:06,693][00704] Avg episode reward: [(0, '17.806')]
299
+ [2024-12-30 06:28:06,695][02840] Saving new best policy, reward=17.806!
300
+ [2024-12-30 06:28:07,649][02853] Updated weights for policy 0, policy_version 670 (0.0013)
301
+ [2024-12-30 06:28:09,638][02853] Updated weights for policy 0, policy_version 680 (0.0013)
302
+ [2024-12-30 06:28:11,628][02853] Updated weights for policy 0, policy_version 690 (0.0012)
303
+ [2024-12-30 06:28:11,690][00704] Fps is (10 sec: 20480.0, 60 sec: 20138.6, 300 sec: 19491.3). Total num frames: 2826240. Throughput: 0: 5035.6. Samples: 706660. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
304
+ [2024-12-30 06:28:11,693][00704] Avg episode reward: [(0, '18.256')]
305
+ [2024-12-30 06:28:11,699][02840] Saving new best policy, reward=18.256!
306
+ [2024-12-30 06:28:13,754][02853] Updated weights for policy 0, policy_version 700 (0.0014)
307
+ [2024-12-30 06:28:15,836][02853] Updated weights for policy 0, policy_version 710 (0.0013)
308
+ [2024-12-30 06:28:16,690][00704] Fps is (10 sec: 20070.3, 60 sec: 20070.4, 300 sec: 19497.0). Total num frames: 2924544. Throughput: 0: 5022.3. Samples: 721154. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
309
+ [2024-12-30 06:28:16,692][00704] Avg episode reward: [(0, '19.306')]
310
+ [2024-12-30 06:28:16,695][02840] Saving new best policy, reward=19.306!
311
+ [2024-12-30 06:28:17,845][02853] Updated weights for policy 0, policy_version 720 (0.0013)
312
+ [2024-12-30 06:28:19,832][02853] Updated weights for policy 0, policy_version 730 (0.0012)
313
+ [2024-12-30 06:28:21,690][00704] Fps is (10 sec: 20070.5, 60 sec: 20138.7, 300 sec: 19528.7). Total num frames: 3026944. Throughput: 0: 5044.6. Samples: 751702. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
314
+ [2024-12-30 06:28:21,692][00704] Avg episode reward: [(0, '22.437')]
315
+ [2024-12-30 06:28:21,700][02840] Saving new best policy, reward=22.437!
316
+ [2024-12-30 06:28:21,809][02853] Updated weights for policy 0, policy_version 740 (0.0014)
317
+ [2024-12-30 06:28:23,808][02853] Updated weights for policy 0, policy_version 750 (0.0013)
318
+ [2024-12-30 06:28:25,801][02853] Updated weights for policy 0, policy_version 760 (0.0013)
319
+ [2024-12-30 06:28:26,690][00704] Fps is (10 sec: 20479.9, 60 sec: 20138.7, 300 sec: 19558.4). Total num frames: 3129344. Throughput: 0: 5046.8. Samples: 782332. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0)
320
+ [2024-12-30 06:28:26,693][00704] Avg episode reward: [(0, '22.555')]
321
+ [2024-12-30 06:28:26,695][02840] Saving new best policy, reward=22.555!
322
+ [2024-12-30 06:28:27,931][02853] Updated weights for policy 0, policy_version 770 (0.0013)
323
+ [2024-12-30 06:28:29,998][02853] Updated weights for policy 0, policy_version 780 (0.0013)
324
+ [2024-12-30 06:28:31,690][00704] Fps is (10 sec: 20070.2, 60 sec: 20138.7, 300 sec: 19561.5). Total num frames: 3227648. Throughput: 0: 5035.5. Samples: 796952. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
325
+ [2024-12-30 06:28:31,693][00704] Avg episode reward: [(0, '20.847')]
326
+ [2024-12-30 06:28:31,991][02853] Updated weights for policy 0, policy_version 790 (0.0013)
327
+ [2024-12-30 06:28:33,977][02853] Updated weights for policy 0, policy_version 800 (0.0013)
328
+ [2024-12-30 06:28:36,010][02853] Updated weights for policy 0, policy_version 810 (0.0013)
329
+ [2024-12-30 06:28:36,690][00704] Fps is (10 sec: 20070.3, 60 sec: 20206.9, 300 sec: 19588.5). Total num frames: 3330048. Throughput: 0: 5049.4. Samples: 827558. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
330
+ [2024-12-30 06:28:36,692][00704] Avg episode reward: [(0, '20.993')]
331
+ [2024-12-30 06:28:38,008][02853] Updated weights for policy 0, policy_version 820 (0.0013)
332
+ [2024-12-30 06:28:39,995][02853] Updated weights for policy 0, policy_version 830 (0.0013)
333
+ [2024-12-30 06:28:41,690][00704] Fps is (10 sec: 20480.3, 60 sec: 20207.0, 300 sec: 19614.0). Total num frames: 3432448. Throughput: 0: 5044.9. Samples: 857936. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
334
+ [2024-12-30 06:28:41,692][00704] Avg episode reward: [(0, '21.620')]
335
+ [2024-12-30 06:28:42,109][02853] Updated weights for policy 0, policy_version 840 (0.0013)
336
+ [2024-12-30 06:28:44,188][02853] Updated weights for policy 0, policy_version 850 (0.0013)
337
+ [2024-12-30 06:28:46,199][02853] Updated weights for policy 0, policy_version 860 (0.0013)
338
+ [2024-12-30 06:28:46,690][00704] Fps is (10 sec: 20070.7, 60 sec: 20138.7, 300 sec: 19615.3). Total num frames: 3530752. Throughput: 0: 5042.4. Samples: 872718. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
339
+ [2024-12-30 06:28:46,693][00704] Avg episode reward: [(0, '21.922')]
340
+ [2024-12-30 06:28:48,219][02853] Updated weights for policy 0, policy_version 870 (0.0013)
341
+ [2024-12-30 06:28:50,205][02853] Updated weights for policy 0, policy_version 880 (0.0012)
342
+ [2024-12-30 06:28:51,690][00704] Fps is (10 sec: 20070.3, 60 sec: 20207.0, 300 sec: 19638.7). Total num frames: 3633152. Throughput: 0: 5056.4. Samples: 903428. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
343
+ [2024-12-30 06:28:51,693][00704] Avg episode reward: [(0, '19.679')]
344
+ [2024-12-30 06:28:52,207][02853] Updated weights for policy 0, policy_version 890 (0.0013)
345
+ [2024-12-30 06:28:54,215][02853] Updated weights for policy 0, policy_version 900 (0.0013)
346
+ [2024-12-30 06:28:56,289][02853] Updated weights for policy 0, policy_version 910 (0.0013)
347
+ [2024-12-30 06:28:56,690][00704] Fps is (10 sec: 20070.5, 60 sec: 20138.7, 300 sec: 19639.2). Total num frames: 3731456. Throughput: 0: 5044.7. Samples: 933670. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
348
+ [2024-12-30 06:28:56,691][00704] Avg episode reward: [(0, '20.748')]
349
+ [2024-12-30 06:28:58,341][02853] Updated weights for policy 0, policy_version 920 (0.0013)
350
+ [2024-12-30 06:29:00,357][02853] Updated weights for policy 0, policy_version 930 (0.0012)
351
+ [2024-12-30 06:29:01,690][00704] Fps is (10 sec: 20070.3, 60 sec: 20206.9, 300 sec: 19660.8). Total num frames: 3833856. Throughput: 0: 5057.5. Samples: 948744. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
352
+ [2024-12-30 06:29:01,693][00704] Avg episode reward: [(0, '20.265')]
353
+ [2024-12-30 06:29:02,347][02853] Updated weights for policy 0, policy_version 940 (0.0013)
354
+ [2024-12-30 06:29:04,344][02853] Updated weights for policy 0, policy_version 950 (0.0013)
355
+ [2024-12-30 06:29:06,348][02853] Updated weights for policy 0, policy_version 960 (0.0013)
356
+ [2024-12-30 06:29:06,690][00704] Fps is (10 sec: 20479.7, 60 sec: 20206.9, 300 sec: 19681.3). Total num frames: 3936256. Throughput: 0: 5062.6. Samples: 979520. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
357
+ [2024-12-30 06:29:06,693][00704] Avg episode reward: [(0, '24.588')]
358
+ [2024-12-30 06:29:06,695][02840] Saving new best policy, reward=24.588!
359
+ [2024-12-30 06:29:08,365][02853] Updated weights for policy 0, policy_version 970 (0.0013)
360
+ [2024-12-30 06:29:10,043][02840] Stopping Batcher_0...
361
+ [2024-12-30 06:29:10,044][02840] Loop batcher_evt_loop terminating...
362
+ [2024-12-30 06:29:10,044][02840] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
363
+ [2024-12-30 06:29:10,043][00704] Component Batcher_0 stopped!
364
+ [2024-12-30 06:29:10,063][02853] Weights refcount: 2 0
365
+ [2024-12-30 06:29:10,065][02853] Stopping InferenceWorker_p0-w0...
366
+ [2024-12-30 06:29:10,065][02853] Loop inference_proc0-0_evt_loop terminating...
367
+ [2024-12-30 06:29:10,065][00704] Component InferenceWorker_p0-w0 stopped!
368
+ [2024-12-30 06:29:10,092][02854] Stopping RolloutWorker_w0...
369
+ [2024-12-30 06:29:10,092][02854] Loop rollout_proc0_evt_loop terminating...
370
+ [2024-12-30 06:29:10,092][00704] Component RolloutWorker_w0 stopped!
371
+ [2024-12-30 06:29:10,096][02858] Stopping RolloutWorker_w4...
372
+ [2024-12-30 06:29:10,097][02858] Loop rollout_proc4_evt_loop terminating...
373
+ [2024-12-30 06:29:10,098][02859] Stopping RolloutWorker_w5...
374
+ [2024-12-30 06:29:10,098][02857] Stopping RolloutWorker_w3...
375
+ [2024-12-30 06:29:10,097][00704] Component RolloutWorker_w4 stopped!
376
+ [2024-12-30 06:29:10,098][02859] Loop rollout_proc5_evt_loop terminating...
377
+ [2024-12-30 06:29:10,099][02857] Loop rollout_proc3_evt_loop terminating...
378
+ [2024-12-30 06:29:10,099][02860] Stopping RolloutWorker_w6...
379
+ [2024-12-30 06:29:10,100][02860] Loop rollout_proc6_evt_loop terminating...
380
+ [2024-12-30 06:29:10,101][02855] Stopping RolloutWorker_w1...
381
+ [2024-12-30 06:29:10,101][02855] Loop rollout_proc1_evt_loop terminating...
382
+ [2024-12-30 06:29:10,102][02861] Stopping RolloutWorker_w7...
383
+ [2024-12-30 06:29:10,099][00704] Component RolloutWorker_w5 stopped!
384
+ [2024-12-30 06:29:10,102][02861] Loop rollout_proc7_evt_loop terminating...
385
+ [2024-12-30 06:29:10,103][02856] Stopping RolloutWorker_w2...
386
+ [2024-12-30 06:29:10,104][02856] Loop rollout_proc2_evt_loop terminating...
387
+ [2024-12-30 06:29:10,103][00704] Component RolloutWorker_w3 stopped!
388
+ [2024-12-30 06:29:10,106][00704] Component RolloutWorker_w6 stopped!
389
+ [2024-12-30 06:29:10,107][00704] Component RolloutWorker_w1 stopped!
390
+ [2024-12-30 06:29:10,108][00704] Component RolloutWorker_w7 stopped!
391
+ [2024-12-30 06:29:10,110][00704] Component RolloutWorker_w2 stopped!
392
+ [2024-12-30 06:29:10,123][02840] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
393
+ [2024-12-30 06:29:10,249][02840] Stopping LearnerWorker_p0...
394
+ [2024-12-30 06:29:10,249][02840] Loop learner_proc0_evt_loop terminating...
395
+ [2024-12-30 06:29:10,249][00704] Component LearnerWorker_p0 stopped!
396
+ [2024-12-30 06:29:10,255][00704] Waiting for process learner_proc0 to stop...
397
+ [2024-12-30 06:29:11,162][00704] Waiting for process inference_proc0-0 to join...
398
+ [2024-12-30 06:29:11,164][00704] Waiting for process rollout_proc0 to join...
399
+ [2024-12-30 06:29:11,166][00704] Waiting for process rollout_proc1 to join...
400
+ [2024-12-30 06:29:11,168][00704] Waiting for process rollout_proc2 to join...
401
+ [2024-12-30 06:29:11,170][00704] Waiting for process rollout_proc3 to join...
402
+ [2024-12-30 06:29:11,172][00704] Waiting for process rollout_proc4 to join...
403
+ [2024-12-30 06:29:11,174][00704] Waiting for process rollout_proc5 to join...
404
+ [2024-12-30 06:29:11,177][00704] Waiting for process rollout_proc6 to join...
405
+ [2024-12-30 06:29:11,179][00704] Waiting for process rollout_proc7 to join...
406
+ [2024-12-30 06:29:11,182][00704] Batcher 0 profile tree view:
407
+ batching: 14.2289, releasing_batches: 0.0245
408
+ [2024-12-30 06:29:11,183][00704] InferenceWorker_p0-w0 profile tree view:
409
+ wait_policy: 0.0001
410
+ wait_policy_total: 3.8212
411
+ update_model: 3.4245
412
+ weight_update: 0.0013
413
+ one_step: 0.0030
414
+ handle_policy_step: 186.4995
415
+ deserialize: 7.6981, stack: 1.2701, obs_to_device_normalize: 46.0042, forward: 87.4663, send_messages: 13.4210
416
+ prepare_outputs: 22.2312
417
+ to_cpu: 13.8908
418
+ [2024-12-30 06:29:11,185][00704] Learner 0 profile tree view:
419
+ misc: 0.0048, prepare_batch: 7.0333
420
+ train: 19.0639
421
+ epoch_init: 0.0054, minibatch_init: 0.0061, losses_postprocess: 0.3146, kl_divergence: 0.3569, after_optimizer: 1.8173
422
+ calculate_losses: 8.7613
423
+ losses_init: 0.0035, forward_head: 0.6262, bptt_initial: 4.8752, tail: 0.6343, advantages_returns: 0.1538, losses: 1.1934
424
+ bptt: 1.1038
425
+ bptt_forward_core: 1.0510
426
+ update: 7.4510
427
+ clip: 0.7858
428
+ [2024-12-30 06:29:11,186][00704] RolloutWorker_w0 profile tree view:
429
+ wait_for_trajectories: 0.1466, enqueue_policy_requests: 7.6354, env_step: 126.0137, overhead: 6.1451, complete_rollouts: 0.2313
430
+ save_policy_outputs: 8.8594
431
+ split_output_tensors: 3.5218
432
+ [2024-12-30 06:29:11,187][00704] RolloutWorker_w7 profile tree view:
433
+ wait_for_trajectories: 0.1423, enqueue_policy_requests: 7.5756, env_step: 126.6795, overhead: 6.1321, complete_rollouts: 0.2277
434
+ save_policy_outputs: 8.9388
435
+ split_output_tensors: 3.5519
436
+ [2024-12-30 06:29:11,189][00704] Loop Runner_EvtLoop terminating...
437
+ [2024-12-30 06:29:11,191][00704] Runner profile tree view:
438
+ main_loop: 214.2976
439
+ [2024-12-30 06:29:11,192][00704] Collected {0: 4005888}, FPS: 18693.1
440
+ [2024-12-30 06:29:30,844][00704] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
441
+ [2024-12-30 06:29:30,846][00704] Overriding arg 'num_workers' with value 1 passed from command line
442
+ [2024-12-30 06:29:30,847][00704] Adding new argument 'no_render'=True that is not in the saved config file!
443
+ [2024-12-30 06:29:30,848][00704] Adding new argument 'save_video'=True that is not in the saved config file!
444
+ [2024-12-30 06:29:30,850][00704] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
445
+ [2024-12-30 06:29:30,851][00704] Adding new argument 'video_name'=None that is not in the saved config file!
446
+ [2024-12-30 06:29:30,853][00704] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file!
447
+ [2024-12-30 06:29:30,854][00704] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
448
+ [2024-12-30 06:29:30,855][00704] Adding new argument 'push_to_hub'=False that is not in the saved config file!
449
+ [2024-12-30 06:29:30,857][00704] Adding new argument 'hf_repository'=None that is not in the saved config file!
450
+ [2024-12-30 06:29:30,858][00704] Adding new argument 'policy_index'=0 that is not in the saved config file!
451
+ [2024-12-30 06:29:30,860][00704] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
452
+ [2024-12-30 06:29:30,861][00704] Adding new argument 'train_script'=None that is not in the saved config file!
453
+ [2024-12-30 06:29:30,863][00704] Adding new argument 'enjoy_script'=None that is not in the saved config file!
454
+ [2024-12-30 06:29:30,865][00704] Using frameskip 1 and render_action_repeat=4 for evaluation
455
+ [2024-12-30 06:29:30,893][00704] Doom resolution: 160x120, resize resolution: (128, 72)
456
+ [2024-12-30 06:29:30,897][00704] RunningMeanStd input shape: (3, 72, 128)
457
+ [2024-12-30 06:29:30,899][00704] RunningMeanStd input shape: (1,)
458
+ [2024-12-30 06:29:30,916][00704] ConvEncoder: input_channels=3
459
+ [2024-12-30 06:29:31,046][00704] Conv encoder output size: 512
460
+ [2024-12-30 06:29:31,048][00704] Policy head output size: 512
461
+ [2024-12-30 06:29:31,210][00704] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
462
+ [2024-12-30 06:29:32,009][00704] Num frames 100...
463
+ [2024-12-30 06:29:32,125][00704] Num frames 200...
464
+ [2024-12-30 06:29:32,248][00704] Num frames 300...
465
+ [2024-12-30 06:29:32,370][00704] Num frames 400...
466
+ [2024-12-30 06:29:32,492][00704] Num frames 500...
467
+ [2024-12-30 06:29:32,620][00704] Num frames 600...
468
+ [2024-12-30 06:29:32,744][00704] Num frames 700...
469
+ [2024-12-30 06:29:32,873][00704] Num frames 800...
470
+ [2024-12-30 06:29:33,002][00704] Num frames 900...
471
+ [2024-12-30 06:29:33,092][00704] Avg episode rewards: #0: 19.280, true rewards: #0: 9.280
472
+ [2024-12-30 06:29:33,094][00704] Avg episode reward: 19.280, avg true_objective: 9.280
473
+ [2024-12-30 06:29:33,185][00704] Num frames 1000...
474
+ [2024-12-30 06:29:33,310][00704] Num frames 1100...
475
+ [2024-12-30 06:29:33,431][00704] Num frames 1200...
476
+ [2024-12-30 06:29:33,550][00704] Num frames 1300...
477
+ [2024-12-30 06:29:33,673][00704] Num frames 1400...
478
+ [2024-12-30 06:29:33,794][00704] Num frames 1500...
479
+ [2024-12-30 06:29:33,910][00704] Num frames 1600...
480
+ [2024-12-30 06:29:34,036][00704] Num frames 1700...
481
+ [2024-12-30 06:29:34,155][00704] Num frames 1800...
482
+ [2024-12-30 06:29:34,273][00704] Num frames 1900...
483
+ [2024-12-30 06:29:34,399][00704] Avg episode rewards: #0: 22.790, true rewards: #0: 9.790
484
+ [2024-12-30 06:29:34,400][00704] Avg episode reward: 22.790, avg true_objective: 9.790
485
+ [2024-12-30 06:29:34,453][00704] Num frames 2000...
486
+ [2024-12-30 06:29:34,573][00704] Num frames 2100...
487
+ [2024-12-30 06:29:34,694][00704] Num frames 2200...
488
+ [2024-12-30 06:29:34,811][00704] Num frames 2300...
489
+ [2024-12-30 06:29:34,931][00704] Num frames 2400...
490
+ [2024-12-30 06:29:35,055][00704] Num frames 2500...
491
+ [2024-12-30 06:29:35,174][00704] Num frames 2600...
492
+ [2024-12-30 06:29:35,291][00704] Num frames 2700...
493
+ [2024-12-30 06:29:35,411][00704] Num frames 2800...
494
+ [2024-12-30 06:29:35,531][00704] Num frames 2900...
495
+ [2024-12-30 06:29:35,652][00704] Num frames 3000...
496
+ [2024-12-30 06:29:35,818][00704] Avg episode rewards: #0: 23.640, true rewards: #0: 10.307
497
+ [2024-12-30 06:29:35,820][00704] Avg episode reward: 23.640, avg true_objective: 10.307
498
+ [2024-12-30 06:29:35,831][00704] Num frames 3100...
499
+ [2024-12-30 06:29:35,948][00704] Num frames 3200...
500
+ [2024-12-30 06:29:36,073][00704] Num frames 3300...
501
+ [2024-12-30 06:29:36,195][00704] Num frames 3400...
502
+ [2024-12-30 06:29:36,314][00704] Num frames 3500...
503
+ [2024-12-30 06:29:36,433][00704] Num frames 3600...
504
+ [2024-12-30 06:29:36,553][00704] Num frames 3700...
505
+ [2024-12-30 06:29:36,670][00704] Num frames 3800...
506
+ [2024-12-30 06:29:36,788][00704] Num frames 3900...
507
+ [2024-12-30 06:29:36,909][00704] Num frames 4000...
508
+ [2024-12-30 06:29:36,988][00704] Avg episode rewards: #0: 23.300, true rewards: #0: 10.050
509
+ [2024-12-30 06:29:36,990][00704] Avg episode reward: 23.300, avg true_objective: 10.050
510
+ [2024-12-30 06:29:37,086][00704] Num frames 4100...
511
+ [2024-12-30 06:29:37,205][00704] Num frames 4200...
512
+ [2024-12-30 06:29:37,323][00704] Num frames 4300...
513
+ [2024-12-30 06:29:37,439][00704] Num frames 4400...
514
+ [2024-12-30 06:29:37,499][00704] Avg episode rewards: #0: 19.408, true rewards: #0: 8.808
515
+ [2024-12-30 06:29:37,500][00704] Avg episode reward: 19.408, avg true_objective: 8.808
516
+ [2024-12-30 06:29:37,614][00704] Num frames 4500...
517
+ [2024-12-30 06:29:37,732][00704] Num frames 4600...
518
+ [2024-12-30 06:29:37,848][00704] Num frames 4700...
519
+ [2024-12-30 06:29:37,963][00704] Num frames 4800...
520
+ [2024-12-30 06:29:38,087][00704] Num frames 4900...
521
+ [2024-12-30 06:29:38,206][00704] Num frames 5000...
522
+ [2024-12-30 06:29:38,326][00704] Num frames 5100...
523
+ [2024-12-30 06:29:38,447][00704] Num frames 5200...
524
+ [2024-12-30 06:29:38,565][00704] Num frames 5300...
525
+ [2024-12-30 06:29:38,686][00704] Num frames 5400...
526
+ [2024-12-30 06:29:38,813][00704] Avg episode rewards: #0: 19.600, true rewards: #0: 9.100
527
+ [2024-12-30 06:29:38,815][00704] Avg episode reward: 19.600, avg true_objective: 9.100
528
+ [2024-12-30 06:29:38,864][00704] Num frames 5500...
529
+ [2024-12-30 06:29:38,981][00704] Num frames 5600...
530
+ [2024-12-30 06:29:39,100][00704] Num frames 5700...
531
+ [2024-12-30 06:29:39,222][00704] Num frames 5800...
532
+ [2024-12-30 06:29:39,339][00704] Num frames 5900...
533
+ [2024-12-30 06:29:39,454][00704] Num frames 6000...
534
+ [2024-12-30 06:29:39,575][00704] Num frames 6100...
535
+ [2024-12-30 06:29:39,695][00704] Num frames 6200...
536
+ [2024-12-30 06:29:39,817][00704] Num frames 6300...
537
+ [2024-12-30 06:29:39,939][00704] Num frames 6400...
538
+ [2024-12-30 06:29:40,059][00704] Num frames 6500...
539
+ [2024-12-30 06:29:40,178][00704] Num frames 6600...
540
+ [2024-12-30 06:29:40,296][00704] Num frames 6700...
541
+ [2024-12-30 06:29:40,440][00704] Avg episode rewards: #0: 21.103, true rewards: #0: 9.674
542
+ [2024-12-30 06:29:40,442][00704] Avg episode reward: 21.103, avg true_objective: 9.674
543
+ [2024-12-30 06:29:40,477][00704] Num frames 6800...
544
+ [2024-12-30 06:29:40,594][00704] Num frames 6900...
545
+ [2024-12-30 06:29:40,712][00704] Num frames 7000...
546
+ [2024-12-30 06:29:40,828][00704] Num frames 7100...
547
+ [2024-12-30 06:29:40,947][00704] Num frames 7200...
548
+ [2024-12-30 06:29:41,068][00704] Num frames 7300...
549
+ [2024-12-30 06:29:41,189][00704] Num frames 7400...
550
+ [2024-12-30 06:29:41,308][00704] Num frames 7500...
551
+ [2024-12-30 06:29:41,426][00704] Num frames 7600...
552
+ [2024-12-30 06:29:41,546][00704] Num frames 7700...
553
+ [2024-12-30 06:29:41,669][00704] Num frames 7800...
554
+ [2024-12-30 06:29:41,805][00704] Avg episode rewards: #0: 21.584, true rewards: #0: 9.834
555
+ [2024-12-30 06:29:41,807][00704] Avg episode reward: 21.584, avg true_objective: 9.834
556
+ [2024-12-30 06:29:41,850][00704] Num frames 7900...
557
+ [2024-12-30 06:29:41,970][00704] Num frames 8000...
558
+ [2024-12-30 06:29:42,091][00704] Num frames 8100...
559
+ [2024-12-30 06:29:42,212][00704] Num frames 8200...
560
+ [2024-12-30 06:29:42,333][00704] Num frames 8300...
561
+ [2024-12-30 06:29:42,452][00704] Num frames 8400...
562
+ [2024-12-30 06:29:42,575][00704] Num frames 8500...
563
+ [2024-12-30 06:29:42,696][00704] Num frames 8600...
564
+ [2024-12-30 06:29:42,832][00704] Avg episode rewards: #0: 20.852, true rewards: #0: 9.630
565
+ [2024-12-30 06:29:42,833][00704] Avg episode reward: 20.852, avg true_objective: 9.630
566
+ [2024-12-30 06:29:42,876][00704] Num frames 8700...
567
+ [2024-12-30 06:29:43,001][00704] Num frames 8800...
568
+ [2024-12-30 06:29:43,125][00704] Num frames 8900...
569
+ [2024-12-30 06:29:43,252][00704] Num frames 9000...
570
+ [2024-12-30 06:29:43,382][00704] Num frames 9100...
571
+ [2024-12-30 06:29:43,505][00704] Num frames 9200...
572
+ [2024-12-30 06:29:43,629][00704] Num frames 9300...
573
+ [2024-12-30 06:29:43,754][00704] Num frames 9400...
574
+ [2024-12-30 06:29:43,878][00704] Num frames 9500...
575
+ [2024-12-30 06:29:44,002][00704] Num frames 9600...
576
+ [2024-12-30 06:29:44,126][00704] Num frames 9700...
577
+ [2024-12-30 06:29:44,263][00704] Num frames 9800...
578
+ [2024-12-30 06:29:44,390][00704] Num frames 9900...
579
+ [2024-12-30 06:29:44,515][00704] Num frames 10000...
580
+ [2024-12-30 06:29:44,641][00704] Num frames 10100...
581
+ [2024-12-30 06:29:44,769][00704] Num frames 10200...
582
+ [2024-12-30 06:29:44,895][00704] Num frames 10300...
583
+ [2024-12-30 06:29:45,067][00704] Avg episode rewards: #0: 23.295, true rewards: #0: 10.395
584
+ [2024-12-30 06:29:45,069][00704] Avg episode reward: 23.295, avg true_objective: 10.395
585
+ [2024-12-30 06:30:09,696][00704] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
586
+ [2024-12-30 06:31:27,179][00704] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
587
+ [2024-12-30 06:31:27,180][00704] Overriding arg 'num_workers' with value 1 passed from command line
588
+ [2024-12-30 06:31:27,181][00704] Adding new argument 'no_render'=True that is not in the saved config file!
589
+ [2024-12-30 06:31:27,183][00704] Adding new argument 'save_video'=True that is not in the saved config file!
590
+ [2024-12-30 06:31:27,184][00704] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
591
+ [2024-12-30 06:31:27,185][00704] Adding new argument 'video_name'=None that is not in the saved config file!
592
+ [2024-12-30 06:31:27,186][00704] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
593
+ [2024-12-30 06:31:27,188][00704] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
594
+ [2024-12-30 06:31:27,189][00704] Adding new argument 'push_to_hub'=True that is not in the saved config file!
595
+ [2024-12-30 06:31:27,190][00704] Adding new argument 'hf_repository'='amanoyaku/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
596
+ [2024-12-30 06:31:27,191][00704] Adding new argument 'policy_index'=0 that is not in the saved config file!
597
+ [2024-12-30 06:31:27,193][00704] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
598
+ [2024-12-30 06:31:27,194][00704] Adding new argument 'train_script'=None that is not in the saved config file!
599
+ [2024-12-30 06:31:27,195][00704] Adding new argument 'enjoy_script'=None that is not in the saved config file!
600
+ [2024-12-30 06:31:27,196][00704] Using frameskip 1 and render_action_repeat=4 for evaluation
601
+ [2024-12-30 06:31:27,220][00704] RunningMeanStd input shape: (3, 72, 128)
602
+ [2024-12-30 06:31:27,223][00704] RunningMeanStd input shape: (1,)
603
+ [2024-12-30 06:31:27,234][00704] ConvEncoder: input_channels=3
604
+ [2024-12-30 06:31:27,272][00704] Conv encoder output size: 512
605
+ [2024-12-30 06:31:27,274][00704] Policy head output size: 512
606
+ [2024-12-30 06:31:27,294][00704] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
607
+ [2024-12-30 06:31:27,711][00704] Num frames 100...
608
+ [2024-12-30 06:31:27,829][00704] Num frames 200...
609
+ [2024-12-30 06:31:27,944][00704] Num frames 300...
610
+ [2024-12-30 06:31:28,071][00704] Num frames 400...
611
+ [2024-12-30 06:31:28,187][00704] Num frames 500...
612
+ [2024-12-30 06:31:28,305][00704] Num frames 600...
613
+ [2024-12-30 06:31:28,420][00704] Num frames 700...
614
+ [2024-12-30 06:31:28,537][00704] Num frames 800...
615
+ [2024-12-30 06:31:28,655][00704] Num frames 900...
616
+ [2024-12-30 06:31:28,781][00704] Num frames 1000...
617
+ [2024-12-30 06:31:28,907][00704] Num frames 1100...
618
+ [2024-12-30 06:31:29,030][00704] Num frames 1200...
619
+ [2024-12-30 06:31:29,150][00704] Num frames 1300...
620
+ [2024-12-30 06:31:29,268][00704] Num frames 1400...
621
+ [2024-12-30 06:31:29,427][00704] Avg episode rewards: #0: 36.820, true rewards: #0: 14.820
622
+ [2024-12-30 06:31:29,429][00704] Avg episode reward: 36.820, avg true_objective: 14.820
623
+ [2024-12-30 06:31:29,453][00704] Num frames 1500...
624
+ [2024-12-30 06:31:29,571][00704] Num frames 1600...
625
+ [2024-12-30 06:31:29,692][00704] Num frames 1700...
626
+ [2024-12-30 06:31:29,809][00704] Num frames 1800...
627
+ [2024-12-30 06:31:29,934][00704] Num frames 1900...
628
+ [2024-12-30 06:31:30,060][00704] Num frames 2000...
629
+ [2024-12-30 06:31:30,182][00704] Avg episode rewards: #0: 22.790, true rewards: #0: 10.290
630
+ [2024-12-30 06:31:30,184][00704] Avg episode reward: 22.790, avg true_objective: 10.290
631
+ [2024-12-30 06:31:30,246][00704] Num frames 2100...
632
+ [2024-12-30 06:31:30,367][00704] Num frames 2200...
633
+ [2024-12-30 06:31:30,484][00704] Num frames 2300...
634
+ [2024-12-30 06:31:30,604][00704] Num frames 2400...
635
+ [2024-12-30 06:31:30,725][00704] Num frames 2500...
636
+ [2024-12-30 06:31:30,843][00704] Num frames 2600...
637
+ [2024-12-30 06:31:30,960][00704] Num frames 2700...
638
+ [2024-12-30 06:31:31,083][00704] Num frames 2800...
639
+ [2024-12-30 06:31:31,203][00704] Num frames 2900...
640
+ [2024-12-30 06:31:31,321][00704] Num frames 3000...
641
+ [2024-12-30 06:31:31,436][00704] Num frames 3100...
642
+ [2024-12-30 06:31:31,557][00704] Num frames 3200...
643
+ [2024-12-30 06:31:31,682][00704] Num frames 3300...
644
+ [2024-12-30 06:31:31,803][00704] Num frames 3400...
645
+ [2024-12-30 06:31:31,924][00704] Num frames 3500...
646
+ [2024-12-30 06:31:32,048][00704] Num frames 3600...
647
+ [2024-12-30 06:31:32,167][00704] Num frames 3700...
648
+ [2024-12-30 06:31:32,290][00704] Num frames 3800...
649
+ [2024-12-30 06:31:32,413][00704] Num frames 3900...
650
+ [2024-12-30 06:31:32,532][00704] Num frames 4000...
651
+ [2024-12-30 06:31:32,667][00704] Avg episode rewards: #0: 31.226, true rewards: #0: 13.560
652
+ [2024-12-30 06:31:32,668][00704] Avg episode reward: 31.226, avg true_objective: 13.560
653
+ [2024-12-30 06:31:32,709][00704] Num frames 4100...
654
+ [2024-12-30 06:31:32,824][00704] Num frames 4200...
655
+ [2024-12-30 06:31:32,940][00704] Num frames 4300...
656
+ [2024-12-30 06:31:33,059][00704] Num frames 4400...
657
+ [2024-12-30 06:31:33,177][00704] Num frames 4500...
658
+ [2024-12-30 06:31:33,299][00704] Num frames 4600...
659
+ [2024-12-30 06:31:33,417][00704] Num frames 4700...
660
+ [2024-12-30 06:31:33,538][00704] Num frames 4800...
661
+ [2024-12-30 06:31:33,657][00704] Num frames 4900...
662
+ [2024-12-30 06:31:33,782][00704] Num frames 5000...
663
+ [2024-12-30 06:31:33,901][00704] Num frames 5100...
664
+ [2024-12-30 06:31:34,060][00704] Avg episode rewards: #0: 28.970, true rewards: #0: 12.970
665
+ [2024-12-30 06:31:34,062][00704] Avg episode reward: 28.970, avg true_objective: 12.970
666
+ [2024-12-30 06:31:34,079][00704] Num frames 5200...
667
+ [2024-12-30 06:31:34,198][00704] Num frames 5300...
668
+ [2024-12-30 06:31:34,319][00704] Num frames 5400...
669
+ [2024-12-30 06:31:34,432][00704] Num frames 5500...
670
+ [2024-12-30 06:31:34,550][00704] Num frames 5600...
671
+ [2024-12-30 06:31:34,682][00704] Num frames 5700...
672
+ [2024-12-30 06:31:34,735][00704] Avg episode rewards: #0: 25.000, true rewards: #0: 11.400
673
+ [2024-12-30 06:31:34,737][00704] Avg episode reward: 25.000, avg true_objective: 11.400
674
+ [2024-12-30 06:31:34,853][00704] Num frames 5800...
675
+ [2024-12-30 06:31:34,972][00704] Num frames 5900...
676
+ [2024-12-30 06:31:35,092][00704] Num frames 6000...
677
+ [2024-12-30 06:31:35,210][00704] Num frames 6100...
678
+ [2024-12-30 06:31:35,329][00704] Num frames 6200...
679
+ [2024-12-30 06:31:35,445][00704] Num frames 6300...
680
+ [2024-12-30 06:31:35,560][00704] Num frames 6400...
681
+ [2024-12-30 06:31:35,680][00704] Num frames 6500...
682
+ [2024-12-30 06:31:35,772][00704] Avg episode rewards: #0: 23.387, true rewards: #0: 10.887
683
+ [2024-12-30 06:31:35,773][00704] Avg episode reward: 23.387, avg true_objective: 10.887
684
+ [2024-12-30 06:31:35,853][00704] Num frames 6600...
685
+ [2024-12-30 06:31:35,970][00704] Num frames 6700...
686
+ [2024-12-30 06:31:36,088][00704] Num frames 6800...
687
+ [2024-12-30 06:31:36,208][00704] Num frames 6900...
688
+ [2024-12-30 06:31:36,326][00704] Num frames 7000...
689
+ [2024-12-30 06:31:36,442][00704] Num frames 7100...
690
+ [2024-12-30 06:31:36,561][00704] Num frames 7200...
691
+ [2024-12-30 06:31:36,678][00704] Num frames 7300...
692
+ [2024-12-30 06:31:36,794][00704] Num frames 7400...
693
+ [2024-12-30 06:31:36,909][00704] Num frames 7500...
694
+ [2024-12-30 06:31:37,029][00704] Num frames 7600...
695
+ [2024-12-30 06:31:37,148][00704] Num frames 7700...
696
+ [2024-12-30 06:31:37,207][00704] Avg episode rewards: #0: 23.576, true rewards: #0: 11.004
697
+ [2024-12-30 06:31:37,209][00704] Avg episode reward: 23.576, avg true_objective: 11.004
698
+ [2024-12-30 06:31:37,323][00704] Num frames 7800...
699
+ [2024-12-30 06:31:37,439][00704] Num frames 7900...
700
+ [2024-12-30 06:31:37,552][00704] Num frames 8000...
701
+ [2024-12-30 06:31:37,666][00704] Num frames 8100...
702
+ [2024-12-30 06:31:37,778][00704] Num frames 8200...
703
+ [2024-12-30 06:31:37,895][00704] Num frames 8300...
704
+ [2024-12-30 06:31:37,964][00704] Avg episode rewards: #0: 21.764, true rewards: #0: 10.389
705
+ [2024-12-30 06:31:37,966][00704] Avg episode reward: 21.764, avg true_objective: 10.389
706
+ [2024-12-30 06:31:38,076][00704] Num frames 8400...
707
+ [2024-12-30 06:31:38,195][00704] Num frames 8500...
708
+ [2024-12-30 06:31:38,315][00704] Num frames 8600...
709
+ [2024-12-30 06:31:38,431][00704] Num frames 8700...
710
+ [2024-12-30 06:31:38,549][00704] Num frames 8800...
711
+ [2024-12-30 06:31:38,667][00704] Num frames 8900...
712
+ [2024-12-30 06:31:38,744][00704] Avg episode rewards: #0: 20.688, true rewards: #0: 9.910
713
+ [2024-12-30 06:31:38,745][00704] Avg episode reward: 20.688, avg true_objective: 9.910
714
+ [2024-12-30 06:31:38,843][00704] Num frames 9000...
715
+ [2024-12-30 06:31:38,959][00704] Num frames 9100...
716
+ [2024-12-30 06:31:39,081][00704] Num frames 9200...
717
+ [2024-12-30 06:31:39,198][00704] Num frames 9300...
718
+ [2024-12-30 06:31:39,335][00704] Avg episode rewards: #0: 19.167, true rewards: #0: 9.367
719
+ [2024-12-30 06:31:39,336][00704] Avg episode reward: 19.167, avg true_objective: 9.367
720
+ [2024-12-30 06:32:01,375][00704] Replay video saved to /content/train_dir/default_experiment/replay.mp4!