Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+replay.mp4 filter=lfs diff=lfs merge=lfs -text

.summary/0/events.out.tfevents.1693653185.cf44f8e7e020

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7ede1cc4e9ba13215dafdf8204a53a805e13226dd9786da35f87e904c6e0a95
+size 136442

README.md

@@ -0,0 +1,56 @@
+---
+library_name: sample-factory
+tags:
+- deep-reinforcement-learning
+- reinforcement-learning
+- sample-factory
+model-index:
+- name: APPO
+  results:
+  - task:
+      type: reinforcement-learning
+      name: reinforcement-learning
+    dataset:
+      name: doom_health_gathering_supreme
+      type: doom_health_gathering_supreme
+    metrics:
+    - type: mean_reward
+      value: 4.19 +/- 0.68
+      name: mean_reward
+      verified: false
+---
+
+A(n) **APPO** model trained on the **doom_health_gathering_supreme** environment.
+
+This model was trained using Sample-Factory 2.0: https://github.com/alex-petrenko/sample-factory.
+Documentation for how to use Sample-Factory can be found at https://www.samplefactory.dev/
+
+
+## Downloading the model
+
+After installing Sample-Factory, download the model with:
+```
+python -m sample_factory.huggingface.load_from_hub -r dimitarrskv/rl_course_vizdoom_health_gathering_supreme
+```
+
+    
+## Using the model
+
+To run the model after download, use the `enjoy` script corresponding to this environment:
+```
+python -m <path.to.enjoy.module> --algo=APPO --env=doom_health_gathering_supreme --train_dir=./train_dir --experiment=rl_course_vizdoom_health_gathering_supreme
+```
+
+
+You can also upload models to the Hugging Face Hub using the same script with the `--push_to_hub` flag.
+See https://www.samplefactory.dev/10-huggingface/huggingface/ for more details
+    
+## Training with this model
+
+To continue training with this model, use the `train` script corresponding to this environment:
+```
+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
+```
+
+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.
+    

checkpoint_p0/best_000000115_471040_reward_4.691.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da343425da33363f96968390fba0ba90d7df77c2be96d416a580a52b2cdf4422
+size 34928614

checkpoint_p0/checkpoint_000000053_217088.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b585910ac84e10d82b04fec15cd18d1ab234686da8db7931ea6b041150b4d99
+size 34929028

checkpoint_p0/checkpoint_000000124_507904.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d19f095175386c51be865a7b8974338b9efef264d459083278d34a3b491394b5
+size 34929028

config.json

@@ -0,0 +1,142 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "doom_health_gathering_supreme",
+  "experiment": "default_experiment",
+  "train_dir": "/content/train_dir",
+  "restart_behavior": "resume",
+  "device": "gpu",
+  "seed": null,
+  "num_policies": 1,
+  "async_rl": true,
+  "serial_mode": false,
+  "batched_sampling": false,
+  "num_batches_to_accumulate": 2,
+  "worker_num_splits": 2,
+  "policy_workers_per_policy": 1,
+  "max_policy_lag": 1000,
+  "num_workers": 8,
+  "num_envs_per_worker": 4,
+  "batch_size": 1024,
+  "num_batches_per_epoch": 1,
+  "num_epochs": 1,
+  "rollout": 32,
+  "recurrence": 32,
+  "shuffle_minibatches": false,
+  "gamma": 0.99,
+  "reward_scale": 1.0,
+  "reward_clip": 1000.0,
+  "value_bootstrap": false,
+  "normalize_returns": true,
+  "exploration_loss_coeff": 0.001,
+  "value_loss_coeff": 0.5,
+  "kl_loss_coeff": 0.0,
+  "exploration_loss": "symmetric_kl",
+  "gae_lambda": 0.95,
+  "ppo_clip_ratio": 0.1,
+  "ppo_clip_value": 0.2,
+  "with_vtrace": false,
+  "vtrace_rho": 1.0,
+  "vtrace_c": 1.0,
+  "optimizer": "adam",
+  "adam_eps": 1e-06,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.999,
+  "max_grad_norm": 4.0,
+  "learning_rate": 0.0001,
+  "lr_schedule": "constant",
+  "lr_schedule_kl_threshold": 0.008,
+  "lr_adaptive_min": 1e-06,
+  "lr_adaptive_max": 0.01,
+  "obs_subtract_mean": 0.0,
+  "obs_scale": 255.0,
+  "normalize_input": true,
+  "normalize_input_keys": null,
+  "decorrelate_experience_max_seconds": 0,
+  "decorrelate_envs_on_one_worker": true,
+  "actor_worker_gpus": [],
+  "set_workers_cpu_affinity": true,
+  "force_envs_single_thread": false,
+  "default_niceness": 0,
+  "log_to_file": true,
+  "experiment_summaries_interval": 10,
+  "flush_summaries_interval": 30,
+  "stats_avg": 100,
+  "summaries_use_frameskip": true,
+  "heartbeat_interval": 20,
+  "heartbeat_reporting_interval": 600,
+  "train_for_env_steps": 500000,
+  "train_for_seconds": 10000000000,
+  "save_every_sec": 120,
+  "keep_checkpoints": 2,
+  "load_checkpoint_kind": "latest",
+  "save_milestones_sec": -1,
+  "save_best_every_sec": 5,
+  "save_best_metric": "reward",
+  "save_best_after": 100000,
+  "benchmark": false,
+  "encoder_mlp_layers": [
+    512,
+    512
+  ],
+  "encoder_conv_architecture": "convnet_simple",
+  "encoder_conv_mlp_layers": [
+    512
+  ],
+  "use_rnn": true,
+  "rnn_size": 512,
+  "rnn_type": "gru",
+  "rnn_num_layers": 1,
+  "decoder_mlp_layers": [],
+  "nonlinearity": "elu",
+  "policy_initialization": "orthogonal",
+  "policy_init_gain": 1.0,
+  "actor_critic_share_weights": true,
+  "adaptive_stddev": true,
+  "continuous_tanh_scale": 0.0,
+  "initial_stddev": 1.0,
+  "use_env_info_cache": false,
+  "env_gpu_actions": false,
+  "env_gpu_observations": true,
+  "env_frameskip": 4,
+  "env_framestack": 1,
+  "pixel_format": "CHW",
+  "use_record_episode_statistics": false,
+  "with_wandb": false,
+  "wandb_user": null,
+  "wandb_project": "sample_factory",
+  "wandb_group": null,
+  "wandb_job_type": "SF",
+  "wandb_tags": [],
+  "with_pbt": false,
+  "pbt_mix_policies_in_one_env": true,
+  "pbt_period_env_steps": 5000000,
+  "pbt_start_mutation": 20000000,
+  "pbt_replace_fraction": 0.3,
+  "pbt_mutation_rate": 0.15,
+  "pbt_replace_reward_gap": 0.1,
+  "pbt_replace_reward_gap_absolute": 1e-06,
+  "pbt_optimize_gamma": false,
+  "pbt_target_objective": "true_objective",
+  "pbt_perturb_min": 1.1,
+  "pbt_perturb_max": 1.5,
+  "num_agents": -1,
+  "num_humans": 0,
+  "num_bots": -1,
+  "start_bot_difficulty": null,
+  "timelimit": null,
+  "res_w": 128,
+  "res_h": 72,
+  "wide_aspect_ratio": false,
+  "eval_env_frameskip": 1,
+  "fps": 35,
+  "command_line": "--env=doom_health_gathering_supreme --num_workers=8 --num_envs_per_worker=4 --train_for_env_steps=500000",
+  "cli_args": {
+    "env": "doom_health_gathering_supreme",
+    "num_workers": 8,
+    "num_envs_per_worker": 4,
+    "train_for_env_steps": 500000
+  },
+  "git_hash": "unknown",
+  "git_repo_name": "not a git repository"
+}

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bf4761d31c1f53f5dc3989216f23977021081c87192987bcbf583ed685cea99
+size 6244445

sf_log.txt

@@ -0,0 +1,486 @@
+[2023-09-02 11:13:14,450][02307] Saving configuration to /content/train_dir/default_experiment/config.json...
+[2023-09-02 11:13:14,453][02307] Rollout worker 0 uses device cpu
+[2023-09-02 11:13:14,455][02307] Rollout worker 1 uses device cpu
+[2023-09-02 11:13:14,456][02307] Rollout worker 2 uses device cpu
+[2023-09-02 11:13:14,457][02307] Rollout worker 3 uses device cpu
+[2023-09-02 11:13:14,458][02307] Rollout worker 4 uses device cpu
+[2023-09-02 11:13:14,459][02307] Rollout worker 5 uses device cpu
+[2023-09-02 11:13:14,460][02307] Rollout worker 6 uses device cpu
+[2023-09-02 11:13:14,461][02307] Rollout worker 7 uses device cpu
+[2023-09-02 11:13:14,669][02307] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-02 11:13:14,673][02307] InferenceWorker_p0-w0: min num requests: 2
+[2023-09-02 11:13:14,750][02307] Starting all processes...
+[2023-09-02 11:13:14,766][02307] Starting process learner_proc0
+[2023-09-02 11:13:14,857][02307] Starting all processes...
+[2023-09-02 11:13:14,883][02307] Starting process inference_proc0-0
+[2023-09-02 11:13:14,897][02307] Starting process rollout_proc0
+[2023-09-02 11:13:14,898][02307] Starting process rollout_proc1
+[2023-09-02 11:13:14,898][02307] Starting process rollout_proc2
+[2023-09-02 11:13:14,898][02307] Starting process rollout_proc3
+[2023-09-02 11:13:14,899][02307] Starting process rollout_proc4
+[2023-09-02 11:13:14,899][02307] Starting process rollout_proc5
+[2023-09-02 11:13:14,899][02307] Starting process rollout_proc6
+[2023-09-02 11:13:14,899][02307] Starting process rollout_proc7
+[2023-09-02 11:13:31,075][10559] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-02 11:13:31,082][10559] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-09-02 11:13:31,134][10559] Num visible devices: 1
+[2023-09-02 11:13:31,186][10559] Starting seed is not provided
+[2023-09-02 11:13:31,187][10559] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-02 11:13:31,188][10559] Initializing actor-critic model on device cuda:0
+[2023-09-02 11:13:31,189][10559] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-02 11:13:31,192][10559] RunningMeanStd input shape: (1,)
+[2023-09-02 11:13:31,287][10578] Worker 4 uses CPU cores [0]
+[2023-09-02 11:13:31,309][10559] ConvEncoder: input_channels=3
+[2023-09-02 11:13:31,451][10580] Worker 6 uses CPU cores [0]
+[2023-09-02 11:13:31,470][10573] Worker 0 uses CPU cores [0]
+[2023-09-02 11:13:31,518][10576] Worker 3 uses CPU cores [1]
+[2023-09-02 11:13:31,541][10577] Worker 5 uses CPU cores [1]
+[2023-09-02 11:13:31,595][10579] Worker 7 uses CPU cores [1]
+[2023-09-02 11:13:31,598][10574] Worker 1 uses CPU cores [1]
+[2023-09-02 11:13:31,598][10575] Worker 2 uses CPU cores [0]
+[2023-09-02 11:13:31,622][10572] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-02 11:13:31,622][10572] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-09-02 11:13:31,637][10572] Num visible devices: 1
+[2023-09-02 11:13:31,738][10559] Conv encoder output size: 512
+[2023-09-02 11:13:31,738][10559] Policy head output size: 512
+[2023-09-02 11:13:31,786][10559] Created Actor Critic model with architecture:
+[2023-09-02 11:13:31,786][10559] ActorCriticSharedWeights(
+  (obs_normalizer): ObservationNormalizer(
+    (running_mean_std): RunningMeanStdDictInPlace(
+      (running_mean_std): ModuleDict(
+        (obs): RunningMeanStdInPlace()
+      )
+    )
+  )
+  (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
+  (encoder): VizdoomEncoder(
+    (basic_encoder): ConvEncoder(
+      (enc): RecursiveScriptModule(
+        original_name=ConvEncoderImpl
+        (conv_head): RecursiveScriptModule(
+          original_name=Sequential
+          (0): RecursiveScriptModule(original_name=Conv2d)
+          (1): RecursiveScriptModule(original_name=ELU)
+          (2): RecursiveScriptModule(original_name=Conv2d)
+          (3): RecursiveScriptModule(original_name=ELU)
+          (4): RecursiveScriptModule(original_name=Conv2d)
+          (5): RecursiveScriptModule(original_name=ELU)
+        )
+        (mlp_layers): RecursiveScriptModule(
+          original_name=Sequential
+          (0): RecursiveScriptModule(original_name=Linear)
+          (1): RecursiveScriptModule(original_name=ELU)
+        )
+      )
+    )
+  )
+  (core): ModelCoreRNN(
+    (core): GRU(512, 512)
+  )
+  (decoder): MlpDecoder(
+    (mlp): Identity()
+  )
+  (critic_linear): Linear(in_features=512, out_features=1, bias=True)
+  (action_parameterization): ActionParameterizationDefault(
+    (distribution_linear): Linear(in_features=512, out_features=5, bias=True)
+  )
+)
+[2023-09-02 11:13:34,660][02307] Heartbeat connected on Batcher_0
+[2023-09-02 11:13:34,670][02307] Heartbeat connected on InferenceWorker_p0-w0
+[2023-09-02 11:13:34,683][02307] Heartbeat connected on RolloutWorker_w0
+[2023-09-02 11:13:34,688][02307] Heartbeat connected on RolloutWorker_w1
+[2023-09-02 11:13:34,692][02307] Heartbeat connected on RolloutWorker_w2
+[2023-09-02 11:13:34,696][02307] Heartbeat connected on RolloutWorker_w3
+[2023-09-02 11:13:34,700][02307] Heartbeat connected on RolloutWorker_w4
+[2023-09-02 11:13:34,711][02307] Heartbeat connected on RolloutWorker_w5
+[2023-09-02 11:13:34,733][02307] Heartbeat connected on RolloutWorker_w6
+[2023-09-02 11:13:34,752][02307] Heartbeat connected on RolloutWorker_w7
+[2023-09-02 11:13:39,825][10559] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-09-02 11:13:39,826][10559] No checkpoints found
+[2023-09-02 11:13:39,826][10559] Did not load from checkpoint, starting from scratch!
+[2023-09-02 11:13:39,827][10559] Initialized policy 0 weights for model version 0
+[2023-09-02 11:13:39,832][10559] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-02 11:13:39,841][10559] LearnerWorker_p0 finished initialization!
+[2023-09-02 11:13:39,842][02307] Heartbeat connected on LearnerWorker_p0
+[2023-09-02 11:13:39,966][10572] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-02 11:13:39,968][10572] RunningMeanStd input shape: (1,)
+[2023-09-02 11:13:39,986][10572] ConvEncoder: input_channels=3
+[2023-09-02 11:13:40,142][10572] Conv encoder output size: 512
+[2023-09-02 11:13:40,143][10572] Policy head output size: 512
+[2023-09-02 11:13:40,278][02307] Inference worker 0-0 is ready!
+[2023-09-02 11:13:40,279][02307] All inference workers are ready! Signal rollout workers to start!
+[2023-09-02 11:13:40,486][10574] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,484][10576] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,488][10579] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,494][10577] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,682][10575] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,693][10578] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,702][10580] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,704][10573] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:13:40,750][02307] 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)
+[2023-09-02 11:13:41,880][10580] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:41,883][10575] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:42,084][10576] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:42,091][10574] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:42,093][10577] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:42,475][10577] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:42,983][10577] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:43,306][10580] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:43,312][10575] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:43,316][10573] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:43,576][10578] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:43,684][10577] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:44,460][10574] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:44,473][10576] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:45,434][10578] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:45,545][10580] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:45,749][02307] 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)
+[2023-09-02 11:13:45,789][10573] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:46,542][10575] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:47,044][10574] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:47,081][10578] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:47,215][10573] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:47,425][10579] Decorrelating experience for 0 frames...
+[2023-09-02 11:13:47,966][10573] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:48,309][10576] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:49,220][10579] Decorrelating experience for 32 frames...
+[2023-09-02 11:13:49,397][10580] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:50,426][10576] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:50,705][10575] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:50,749][02307] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 121.4. Samples: 1214. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-09-02 11:13:50,754][02307] Avg episode reward: [(0, '3.186')]
+[2023-09-02 11:13:52,449][10559] Signal inference workers to stop experience collection...
+[2023-09-02 11:13:52,464][10572] InferenceWorker_p0-w0: stopping experience collection
+[2023-09-02 11:13:52,668][10574] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:52,820][10578] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:54,082][10579] Decorrelating experience for 64 frames...
+[2023-09-02 11:13:55,122][10579] Decorrelating experience for 96 frames...
+[2023-09-02 11:13:55,748][02307] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 152.0. Samples: 2280. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-09-02 11:13:55,751][02307] Avg episode reward: [(0, '3.058')]
+[2023-09-02 11:13:56,652][10559] Signal inference workers to resume experience collection...
+[2023-09-02 11:13:56,653][10572] InferenceWorker_p0-w0: resuming experience collection
+[2023-09-02 11:14:00,748][02307] Fps is (10 sec: 819.2, 60 sec: 409.6, 300 sec: 409.6). Total num frames: 8192. Throughput: 0: 143.8. Samples: 2876. Policy #0 lag: (min: 1.0, avg: 1.0, max: 1.0)
+[2023-09-02 11:14:00,752][02307] Avg episode reward: [(0, '2.819')]
+[2023-09-02 11:14:05,749][02307] Fps is (10 sec: 2867.2, 60 sec: 1146.9, 300 sec: 1146.9). Total num frames: 28672. Throughput: 0: 284.5. Samples: 7112. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:14:05,760][02307] Avg episode reward: [(0, '3.575')]
+[2023-09-02 11:14:08,166][10572] Updated weights for policy 0, policy_version 10 (0.0013)
+[2023-09-02 11:14:10,753][02307] Fps is (10 sec: 4094.3, 60 sec: 1638.3, 300 sec: 1638.3). Total num frames: 49152. Throughput: 0: 423.9. Samples: 12718. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-09-02 11:14:10,758][02307] Avg episode reward: [(0, '4.120')]
+[2023-09-02 11:14:15,754][02307] Fps is (10 sec: 3274.9, 60 sec: 1755.2, 300 sec: 1755.2). Total num frames: 61440. Throughput: 0: 421.6. Samples: 14758. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
+[2023-09-02 11:14:15,757][02307] Avg episode reward: [(0, '4.419')]
+[2023-09-02 11:14:20,749][02307] Fps is (10 sec: 2458.6, 60 sec: 1843.3, 300 sec: 1843.3). Total num frames: 73728. Throughput: 0: 458.0. Samples: 18318. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-09-02 11:14:20,756][02307] Avg episode reward: [(0, '4.492')]
+[2023-09-02 11:14:22,814][10572] Updated weights for policy 0, policy_version 20 (0.0047)
+[2023-09-02 11:14:25,748][02307] Fps is (10 sec: 2868.9, 60 sec: 2002.6, 300 sec: 2002.6). Total num frames: 90112. Throughput: 0: 518.5. Samples: 23330. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-09-02 11:14:25,756][02307] Avg episode reward: [(0, '4.464')]
+[2023-09-02 11:14:30,748][02307] Fps is (10 sec: 3686.4, 60 sec: 2211.9, 300 sec: 2211.9). Total num frames: 110592. Throughput: 0: 585.3. Samples: 26340. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
+[2023-09-02 11:14:30,751][02307] Avg episode reward: [(0, '4.339')]
+[2023-09-02 11:14:30,754][10559] Saving new best policy, reward=4.339!
+[2023-09-02 11:14:33,815][10572] Updated weights for policy 0, policy_version 30 (0.0031)
+[2023-09-02 11:14:35,749][02307] Fps is (10 sec: 3686.4, 60 sec: 2308.7, 300 sec: 2308.7). Total num frames: 126976. Throughput: 0: 671.9. Samples: 31450. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-09-02 11:14:35,755][02307] Avg episode reward: [(0, '4.350')]
+[2023-09-02 11:14:35,763][10559] Saving new best policy, reward=4.350!
+[2023-09-02 11:14:40,750][02307] Fps is (10 sec: 2866.7, 60 sec: 2321.1, 300 sec: 2321.1). Total num frames: 139264. Throughput: 0: 731.4. Samples: 35196. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
+[2023-09-02 11:14:40,753][02307] Avg episode reward: [(0, '4.580')]
+[2023-09-02 11:14:40,756][10559] Saving new best policy, reward=4.580!
+[2023-09-02 11:14:45,749][02307] Fps is (10 sec: 2867.2, 60 sec: 2594.1, 300 sec: 2394.6). Total num frames: 155648. Throughput: 0: 763.3. Samples: 37226. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
+[2023-09-02 11:14:45,751][02307] Avg episode reward: [(0, '4.562')]
+[2023-09-02 11:14:47,692][10572] Updated weights for policy 0, policy_version 40 (0.0019)
+[2023-09-02 11:14:50,748][02307] Fps is (10 sec: 3277.4, 60 sec: 2867.2, 300 sec: 2457.7). Total num frames: 172032. Throughput: 0: 801.2. Samples: 43164. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:14:50,755][02307] Avg episode reward: [(0, '4.499')]
+[2023-09-02 11:14:55,749][02307] Fps is (10 sec: 3276.8, 60 sec: 3140.3, 300 sec: 2512.3). Total num frames: 188416. Throughput: 0: 778.6. Samples: 47750. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
+[2023-09-02 11:14:55,754][02307] Avg episode reward: [(0, '4.529')]
+[2023-09-02 11:15:00,751][02307] Fps is (10 sec: 2866.4, 60 sec: 3208.4, 300 sec: 2508.8). Total num frames: 200704. Throughput: 0: 775.6. Samples: 49658. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
+[2023-09-02 11:15:00,755][02307] Avg episode reward: [(0, '4.474')]
+[2023-09-02 11:15:01,704][10572] Updated weights for policy 0, policy_version 50 (0.0039)
+[2023-09-02 11:15:05,749][02307] Fps is (10 sec: 2867.2, 60 sec: 3140.3, 300 sec: 2554.0). Total num frames: 217088. Throughput: 0: 781.3. Samples: 53478. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:15:05,754][02307] Avg episode reward: [(0, '4.404')]
+[2023-09-02 11:15:05,762][10559] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000053_217088.pth...
+[2023-09-02 11:15:10,749][02307] Fps is (10 sec: 3277.6, 60 sec: 3072.2, 300 sec: 2594.2). Total num frames: 233472. Throughput: 0: 801.5. Samples: 59398. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:15:10,756][02307] Avg episode reward: [(0, '4.407')]
+[2023-09-02 11:15:12,920][10572] Updated weights for policy 0, policy_version 60 (0.0024)
+[2023-09-02 11:15:15,748][02307] Fps is (10 sec: 3276.8, 60 sec: 3140.6, 300 sec: 2630.1). Total num frames: 249856. Throughput: 0: 799.3. Samples: 62308. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0)
+[2023-09-02 11:15:15,751][02307] Avg episode reward: [(0, '4.386')]
+[2023-09-02 11:15:20,748][02307] Fps is (10 sec: 3276.9, 60 sec: 3208.5, 300 sec: 2662.4). Total num frames: 266240. Throughput: 0: 773.2. Samples: 66246. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:15:20,751][02307] Avg episode reward: [(0, '4.529')]
+[2023-09-02 11:15:25,749][02307] Fps is (10 sec: 2867.2, 60 sec: 3140.3, 300 sec: 2652.7). Total num frames: 278528. Throughput: 0: 774.6. Samples: 70050. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-09-02 11:15:25,756][02307] Avg episode reward: [(0, '4.409')]
+[2023-09-02 11:15:27,634][10572] Updated weights for policy 0, policy_version 70 (0.0043)
+[2023-09-02 11:15:30,748][02307] Fps is (10 sec: 3276.8, 60 sec: 3140.3, 300 sec: 2718.3). Total num frames: 299008. Throughput: 0: 794.4. Samples: 72976. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:15:30,756][02307] Avg episode reward: [(0, '4.445')]
+[2023-09-02 11:15:35,749][02307] Fps is (10 sec: 3686.1, 60 sec: 3140.2, 300 sec: 2742.6). Total num frames: 315392. Throughput: 0: 795.5. Samples: 78960. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0)
+[2023-09-02 11:15:35,752][02307] Avg episode reward: [(0, '4.537')]
+[2023-09-02 11:15:39,050][10572] Updated weights for policy 0, policy_version 80 (0.0026)
+[2023-09-02 11:15:40,748][02307] Fps is (10 sec: 3276.8, 60 sec: 3208.6, 300 sec: 2764.8). Total num frames: 331776. Throughput: 0: 788.0. Samples: 83210. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
+[2023-09-02 11:15:40,754][02307] Avg episode reward: [(0, '4.472')]
+[2023-09-02 11:15:45,750][02307] Fps is (10 sec: 2866.9, 60 sec: 3140.2, 300 sec: 2752.5). Total num frames: 344064. Throughput: 0: 787.5. Samples: 85096. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
+[2023-09-02 11:15:45,753][02307] Avg episode reward: [(0, '4.476')]
+[2023-09-02 11:15:50,748][02307] Fps is (10 sec: 2867.2, 60 sec: 3140.3, 300 sec: 2772.7). Total num frames: 360448. Throughput: 0: 804.7. Samples: 89690. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:15:50,751][02307] Avg episode reward: [(0, '4.482')]
+[2023-09-02 11:15:52,229][10572] Updated weights for policy 0, policy_version 90 (0.0035)
+[2023-09-02 11:15:55,748][02307] Fps is (10 sec: 3687.0, 60 sec: 3208.5, 300 sec: 2821.7). Total num frames: 380928. Throughput: 0: 809.7. Samples: 95834. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:15:55,751][02307] Avg episode reward: [(0, '4.533')]
+[2023-09-02 11:16:00,749][02307] Fps is (10 sec: 3276.8, 60 sec: 3208.7, 300 sec: 2808.7). Total num frames: 393216. Throughput: 0: 796.1. Samples: 98134. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
+[2023-09-02 11:16:00,751][02307] Avg episode reward: [(0, '4.540')]
+[2023-09-02 11:16:05,749][02307] Fps is (10 sec: 2457.5, 60 sec: 3140.2, 300 sec: 2796.6). Total num frames: 405504. Throughput: 0: 792.2. Samples: 101894. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
+[2023-09-02 11:16:05,757][02307] Avg episode reward: [(0, '4.650')]
+[2023-09-02 11:16:05,770][10559] Saving new best policy, reward=4.650!
+[2023-09-02 11:16:06,089][10572] Updated weights for policy 0, policy_version 100 (0.0013)
+[2023-09-02 11:16:10,748][02307] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 2839.9). Total num frames: 425984. Throughput: 0: 813.4. Samples: 106652. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0)
+[2023-09-02 11:16:10,751][02307] Avg episode reward: [(0, '4.556')]
+[2023-09-02 11:16:15,748][02307] Fps is (10 sec: 3686.6, 60 sec: 3208.5, 300 sec: 2854.0). Total num frames: 442368. Throughput: 0: 815.0. Samples: 109650. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0)
+[2023-09-02 11:16:15,755][02307] Avg episode reward: [(0, '4.647')]
+[2023-09-02 11:16:16,843][10572] Updated weights for policy 0, policy_version 110 (0.0022)
+[2023-09-02 11:16:20,748][02307] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 2867.2). Total num frames: 458752. Throughput: 0: 797.8. Samples: 114860. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0)
+[2023-09-02 11:16:20,751][02307] Avg episode reward: [(0, '4.559')]
+[2023-09-02 11:16:25,749][02307] Fps is (10 sec: 2867.0, 60 sec: 3208.5, 300 sec: 2854.8). Total num frames: 471040. Throughput: 0: 790.0. Samples: 118762. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:16:25,755][02307] Avg episode reward: [(0, '4.691')]
+[2023-09-02 11:16:25,769][10559] Saving new best policy, reward=4.691!
+[2023-09-02 11:16:30,750][02307] Fps is (10 sec: 2866.7, 60 sec: 3140.2, 300 sec: 2867.2). Total num frames: 487424. Throughput: 0: 788.4. Samples: 120572. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0)
+[2023-09-02 11:16:30,752][02307] Avg episode reward: [(0, '4.614')]
+[2023-09-02 11:16:31,245][10572] Updated weights for policy 0, policy_version 120 (0.0019)
+[2023-09-02 11:16:35,519][10559] Stopping Batcher_0...
+[2023-09-02 11:16:35,519][10559] Loop batcher_evt_loop terminating...
+[2023-09-02 11:16:35,520][02307] Component Batcher_0 stopped!
+[2023-09-02 11:16:35,530][10559] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000124_507904.pth...
+[2023-09-02 11:16:35,557][10575] Stopping RolloutWorker_w2...
+[2023-09-02 11:16:35,558][02307] Component RolloutWorker_w2 stopped!
+[2023-09-02 11:16:35,562][10575] Loop rollout_proc2_evt_loop terminating...
+[2023-09-02 11:16:35,591][10580] Stopping RolloutWorker_w6...
+[2023-09-02 11:16:35,590][02307] Component RolloutWorker_w6 stopped!
+[2023-09-02 11:16:35,597][10580] Loop rollout_proc6_evt_loop terminating...
+[2023-09-02 11:16:35,610][10578] Stopping RolloutWorker_w4...
+[2023-09-02 11:16:35,609][02307] Component RolloutWorker_w4 stopped!
+[2023-09-02 11:16:35,617][02307] Component RolloutWorker_w7 stopped!
+[2023-09-02 11:16:35,619][10579] Stopping RolloutWorker_w7...
+[2023-09-02 11:16:35,623][10579] Loop rollout_proc7_evt_loop terminating...
+[2023-09-02 11:16:35,610][10578] Loop rollout_proc4_evt_loop terminating...
+[2023-09-02 11:16:35,627][02307] Component RolloutWorker_w3 stopped!
+[2023-09-02 11:16:35,628][10576] Stopping RolloutWorker_w3...
+[2023-09-02 11:16:35,632][02307] Component RolloutWorker_w5 stopped!
+[2023-09-02 11:16:35,634][10577] Stopping RolloutWorker_w5...
+[2023-09-02 11:16:35,638][10577] Loop rollout_proc5_evt_loop terminating...
+[2023-09-02 11:16:35,631][10576] Loop rollout_proc3_evt_loop terminating...
+[2023-09-02 11:16:35,637][10572] Weights refcount: 2 0
+[2023-09-02 11:16:35,641][02307] Component InferenceWorker_p0-w0 stopped!
+[2023-09-02 11:16:35,643][02307] Component RolloutWorker_w0 stopped!
+[2023-09-02 11:16:35,641][10572] Stopping InferenceWorker_p0-w0...
+[2023-09-02 11:16:35,650][10572] Loop inference_proc0-0_evt_loop terminating...
+[2023-09-02 11:16:35,642][10573] Stopping RolloutWorker_w0...
+[2023-09-02 11:16:35,655][02307] Component RolloutWorker_w1 stopped!
+[2023-09-02 11:16:35,657][10574] Stopping RolloutWorker_w1...
+[2023-09-02 11:16:35,659][10574] Loop rollout_proc1_evt_loop terminating...
+[2023-09-02 11:16:35,653][10573] Loop rollout_proc0_evt_loop terminating...
+[2023-09-02 11:16:35,682][10559] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000124_507904.pth...
+[2023-09-02 11:16:35,886][10559] Stopping LearnerWorker_p0...
+[2023-09-02 11:16:35,886][02307] Component LearnerWorker_p0 stopped!
+[2023-09-02 11:16:35,888][02307] Waiting for process learner_proc0 to stop...
+[2023-09-02 11:16:35,888][10559] Loop learner_proc0_evt_loop terminating...
+[2023-09-02 11:16:37,785][02307] Waiting for process inference_proc0-0 to join...
+[2023-09-02 11:16:37,788][02307] Waiting for process rollout_proc0 to join...
+[2023-09-02 11:16:39,731][02307] Waiting for process rollout_proc1 to join...
+[2023-09-02 11:16:39,734][02307] Waiting for process rollout_proc2 to join...
+[2023-09-02 11:16:39,735][02307] Waiting for process rollout_proc3 to join...
+[2023-09-02 11:16:39,737][02307] Waiting for process rollout_proc4 to join...
+[2023-09-02 11:16:39,739][02307] Waiting for process rollout_proc5 to join...
+[2023-09-02 11:16:39,741][02307] Waiting for process rollout_proc6 to join...
+[2023-09-02 11:16:39,743][02307] Waiting for process rollout_proc7 to join...
+[2023-09-02 11:16:39,746][02307] Batcher 0 profile tree view:
+batching: 3.9791, releasing_batches: 0.0058
+[2023-09-02 11:16:39,747][02307] InferenceWorker_p0-w0 profile tree view:
+wait_policy: 0.0043
+  wait_policy_total: 80.1493
+update_model: 1.2177
+  weight_update: 0.0022
+one_step: 0.0039
+  handle_policy_step: 83.9183
+    deserialize: 2.2387, stack: 0.4321, obs_to_device_normalize: 16.0242, forward: 47.0405, send_messages: 3.7519
+    prepare_outputs: 10.5362
+      to_cpu: 5.9112
+[2023-09-02 11:16:39,749][02307] Learner 0 profile tree view:
+misc: 0.0008, prepare_batch: 10.9081
+train: 11.7425
+  epoch_init: 0.0008, minibatch_init: 0.0009, losses_postprocess: 0.0622, kl_divergence: 0.0836, after_optimizer: 0.5661
+  calculate_losses: 3.7279
+    losses_init: 0.0005, forward_head: 0.3784, bptt_initial: 2.3691, tail: 0.1694, advantages_returns: 0.0290, losses: 0.4587
+    bptt: 0.2574
+      bptt_forward_core: 0.2505
+  update: 7.1774
+    clip: 4.1199
+[2023-09-02 11:16:39,751][02307] RolloutWorker_w0 profile tree view:
+wait_for_trajectories: 0.0691, enqueue_policy_requests: 22.7559, env_step: 120.8244, overhead: 3.8799, complete_rollouts: 1.1832
+save_policy_outputs: 3.1243
+  split_output_tensors: 1.5323
+[2023-09-02 11:16:39,752][02307] RolloutWorker_w7 profile tree view:
+wait_for_trajectories: 0.0595, enqueue_policy_requests: 21.1253, env_step: 119.6121, overhead: 3.2637, complete_rollouts: 0.8818
+save_policy_outputs: 3.0154
+  split_output_tensors: 1.4709
+[2023-09-02 11:16:39,754][02307] Loop Runner_EvtLoop terminating...
+[2023-09-02 11:16:39,755][02307] Runner profile tree view:
+main_loop: 205.0056
+[2023-09-02 11:16:39,759][02307] Collected {0: 507904}, FPS: 2477.5
+[2023-09-02 11:16:39,820][02307] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2023-09-02 11:16:39,823][02307] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-09-02 11:16:39,825][02307] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-09-02 11:16:39,829][02307] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-09-02 11:16:39,830][02307] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-09-02 11:16:39,832][02307] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-09-02 11:16:39,833][02307] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file!
+[2023-09-02 11:16:39,835][02307] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-09-02 11:16:39,836][02307] Adding new argument 'push_to_hub'=False that is not in the saved config file!
+[2023-09-02 11:16:39,837][02307] Adding new argument 'hf_repository'=None that is not in the saved config file!
+[2023-09-02 11:16:39,838][02307] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-09-02 11:16:39,839][02307] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-09-02 11:16:39,840][02307] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-09-02 11:16:39,841][02307] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-09-02 11:16:39,842][02307] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-09-02 11:16:39,888][02307] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-02 11:16:39,893][02307] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-02 11:16:39,895][02307] RunningMeanStd input shape: (1,)
+[2023-09-02 11:16:39,917][02307] ConvEncoder: input_channels=3
+[2023-09-02 11:16:40,111][02307] Conv encoder output size: 512
+[2023-09-02 11:16:40,113][02307] Policy head output size: 512
+[2023-09-02 11:16:43,131][02307] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000124_507904.pth...
+[2023-09-02 11:16:44,366][02307] Num frames 100...
+[2023-09-02 11:16:44,493][02307] Num frames 200...
+[2023-09-02 11:16:44,625][02307] Num frames 300...
+[2023-09-02 11:16:44,791][02307] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-09-02 11:16:44,792][02307] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-09-02 11:16:44,819][02307] Num frames 400...
+[2023-09-02 11:16:44,960][02307] Num frames 500...
+[2023-09-02 11:16:45,100][02307] Num frames 600...
+[2023-09-02 11:16:45,224][02307] Num frames 700...
+[2023-09-02 11:16:45,362][02307] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-09-02 11:16:45,363][02307] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-09-02 11:16:45,410][02307] Num frames 800...
+[2023-09-02 11:16:45,549][02307] Num frames 900...
+[2023-09-02 11:16:45,692][02307] Num frames 1000...
+[2023-09-02 11:16:45,843][02307] Num frames 1100...
+[2023-09-02 11:16:45,972][02307] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-09-02 11:16:45,973][02307] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-09-02 11:16:46,049][02307] Num frames 1200...
+[2023-09-02 11:16:46,180][02307] Num frames 1300...
+[2023-09-02 11:16:46,313][02307] Num frames 1400...
+[2023-09-02 11:16:46,440][02307] Num frames 1500...
+[2023-09-02 11:16:46,581][02307] Num frames 1600...
+[2023-09-02 11:16:46,633][02307] Avg episode rewards: #0: 4.250, true rewards: #0: 4.000
+[2023-09-02 11:16:46,634][02307] Avg episode reward: 4.250, avg true_objective: 4.000
+[2023-09-02 11:16:46,768][02307] Num frames 1700...
+[2023-09-02 11:16:46,905][02307] Num frames 1800...
+[2023-09-02 11:16:47,034][02307] Num frames 1900...
+[2023-09-02 11:16:47,168][02307] Num frames 2000...
+[2023-09-02 11:16:47,296][02307] Num frames 2100...
+[2023-09-02 11:16:47,408][02307] Avg episode rewards: #0: 4.888, true rewards: #0: 4.288
+[2023-09-02 11:16:47,410][02307] Avg episode reward: 4.888, avg true_objective: 4.288
+[2023-09-02 11:16:47,496][02307] Num frames 2200...
+[2023-09-02 11:16:47,643][02307] Num frames 2300...
+[2023-09-02 11:16:47,790][02307] Num frames 2400...
+[2023-09-02 11:16:47,922][02307] Num frames 2500...
+[2023-09-02 11:16:48,099][02307] Avg episode rewards: #0: 4.987, true rewards: #0: 4.320
+[2023-09-02 11:16:48,101][02307] Avg episode reward: 4.987, avg true_objective: 4.320
+[2023-09-02 11:16:48,116][02307] Num frames 2600...
+[2023-09-02 11:16:48,252][02307] Num frames 2700...
+[2023-09-02 11:16:48,382][02307] Num frames 2800...
+[2023-09-02 11:16:48,498][02307] Avg episode rewards: #0: 4.640, true rewards: #0: 4.069
+[2023-09-02 11:16:48,501][02307] Avg episode reward: 4.640, avg true_objective: 4.069
+[2023-09-02 11:16:48,579][02307] Num frames 2900...
+[2023-09-02 11:16:48,705][02307] Num frames 3000...
+[2023-09-02 11:16:48,844][02307] Num frames 3100...
+[2023-09-02 11:16:48,976][02307] Num frames 3200...
+[2023-09-02 11:16:49,031][02307] Avg episode rewards: #0: 4.625, true rewards: #0: 4.000
+[2023-09-02 11:16:49,032][02307] Avg episode reward: 4.625, avg true_objective: 4.000
+[2023-09-02 11:16:49,168][02307] Num frames 3300...
+[2023-09-02 11:16:49,296][02307] Num frames 3400...
+[2023-09-02 11:16:49,429][02307] Num frames 3500...
+[2023-09-02 11:16:49,586][02307] Avg episode rewards: #0: 4.538, true rewards: #0: 3.982
+[2023-09-02 11:16:49,588][02307] Avg episode reward: 4.538, avg true_objective: 3.982
+[2023-09-02 11:16:49,613][02307] Num frames 3600...
+[2023-09-02 11:16:49,748][02307] Num frames 3700...
+[2023-09-02 11:16:49,894][02307] Num frames 3800...
+[2023-09-02 11:16:50,019][02307] Num frames 3900...
+[2023-09-02 11:16:50,159][02307] Avg episode rewards: #0: 4.468, true rewards: #0: 3.968
+[2023-09-02 11:16:50,160][02307] Avg episode reward: 4.468, avg true_objective: 3.968
+[2023-09-02 11:17:15,137][02307] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
+[2023-09-02 11:17:15,310][02307] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2023-09-02 11:17:15,312][02307] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-09-02 11:17:15,314][02307] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-09-02 11:17:15,317][02307] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-09-02 11:17:15,319][02307] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-09-02 11:17:15,321][02307] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-09-02 11:17:15,323][02307] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
+[2023-09-02 11:17:15,323][02307] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-09-02 11:17:15,324][02307] Adding new argument 'push_to_hub'=True that is not in the saved config file!
+[2023-09-02 11:17:15,325][02307] Adding new argument 'hf_repository'='dimitarrskv/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
+[2023-09-02 11:17:15,326][02307] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-09-02 11:17:15,327][02307] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-09-02 11:17:15,328][02307] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-09-02 11:17:15,329][02307] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-09-02 11:17:15,330][02307] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-09-02 11:17:15,377][02307] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-02 11:17:15,380][02307] RunningMeanStd input shape: (1,)
+[2023-09-02 11:17:15,397][02307] ConvEncoder: input_channels=3
+[2023-09-02 11:17:15,455][02307] Conv encoder output size: 512
+[2023-09-02 11:17:15,458][02307] Policy head output size: 512
+[2023-09-02 11:17:15,488][02307] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000124_507904.pth...
+[2023-09-02 11:17:16,362][02307] Num frames 100...
+[2023-09-02 11:17:16,541][02307] Num frames 200...
+[2023-09-02 11:17:16,729][02307] Num frames 300...
+[2023-09-02 11:17:16,955][02307] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-09-02 11:17:16,958][02307] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-09-02 11:17:17,006][02307] Num frames 400...
+[2023-09-02 11:17:17,209][02307] Num frames 500...
+[2023-09-02 11:17:17,386][02307] Num frames 600...
+[2023-09-02 11:17:17,574][02307] Num frames 700...
+[2023-09-02 11:17:17,770][02307] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-09-02 11:17:17,772][02307] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-09-02 11:17:17,858][02307] Num frames 800...
+[2023-09-02 11:17:18,076][02307] Num frames 900...
+[2023-09-02 11:17:18,296][02307] Num frames 1000...
+[2023-09-02 11:17:18,515][02307] Num frames 1100...
+[2023-09-02 11:17:18,676][02307] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-09-02 11:17:18,678][02307] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-09-02 11:17:18,810][02307] Num frames 1200...
+[2023-09-02 11:17:19,020][02307] Num frames 1300...
+[2023-09-02 11:17:19,208][02307] Num frames 1400...
+[2023-09-02 11:17:19,397][02307] Num frames 1500...
+[2023-09-02 11:17:19,598][02307] Num frames 1600...
+[2023-09-02 11:17:19,653][02307] Avg episode rewards: #0: 4.250, true rewards: #0: 4.000
+[2023-09-02 11:17:19,655][02307] Avg episode reward: 4.250, avg true_objective: 4.000
+[2023-09-02 11:17:19,848][02307] Num frames 1700...
+[2023-09-02 11:17:20,039][02307] Num frames 1800...
+[2023-09-02 11:17:20,225][02307] Num frames 1900...
+[2023-09-02 11:17:20,416][02307] Num frames 2000...
+[2023-09-02 11:17:20,567][02307] Avg episode rewards: #0: 4.496, true rewards: #0: 4.096
+[2023-09-02 11:17:20,569][02307] Avg episode reward: 4.496, avg true_objective: 4.096
+[2023-09-02 11:17:20,672][02307] Num frames 2100...
+[2023-09-02 11:17:20,858][02307] Num frames 2200...
+[2023-09-02 11:17:21,015][02307] Num frames 2300...
+[2023-09-02 11:17:21,142][02307] Num frames 2400...
+[2023-09-02 11:17:21,269][02307] Num frames 2500...
+[2023-09-02 11:17:21,397][02307] Num frames 2600...
+[2023-09-02 11:17:21,523][02307] Avg episode rewards: #0: 5.260, true rewards: #0: 4.427
+[2023-09-02 11:17:21,525][02307] Avg episode reward: 5.260, avg true_objective: 4.427
+[2023-09-02 11:17:21,586][02307] Num frames 2700...
+[2023-09-02 11:17:21,714][02307] Num frames 2800...
+[2023-09-02 11:17:21,838][02307] Num frames 2900...
+[2023-09-02 11:17:21,962][02307] Num frames 3000...
+[2023-09-02 11:17:22,073][02307] Avg episode rewards: #0: 5.057, true rewards: #0: 4.343
+[2023-09-02 11:17:22,075][02307] Avg episode reward: 5.057, avg true_objective: 4.343
+[2023-09-02 11:17:22,158][02307] Num frames 3100...
+[2023-09-02 11:17:22,289][02307] Num frames 3200...
+[2023-09-02 11:17:22,423][02307] Num frames 3300...
+[2023-09-02 11:17:22,556][02307] Num frames 3400...
+[2023-09-02 11:17:22,643][02307] Avg episode rewards: #0: 4.905, true rewards: #0: 4.280
+[2023-09-02 11:17:22,645][02307] Avg episode reward: 4.905, avg true_objective: 4.280
+[2023-09-02 11:17:22,789][02307] Num frames 3500...
+[2023-09-02 11:17:22,975][02307] Num frames 3600...
+[2023-09-02 11:17:23,166][02307] Num frames 3700...
+[2023-09-02 11:17:23,347][02307] Num frames 3800...
+[2023-09-02 11:17:23,422][02307] Avg episode rewards: #0: 4.787, true rewards: #0: 4.231
+[2023-09-02 11:17:23,426][02307] Avg episode reward: 4.787, avg true_objective: 4.231
+[2023-09-02 11:17:23,608][02307] Num frames 3900...
+[2023-09-02 11:17:23,795][02307] Num frames 4000...
+[2023-09-02 11:17:23,974][02307] Num frames 4100...
+[2023-09-02 11:17:24,198][02307] Avg episode rewards: #0: 4.692, true rewards: #0: 4.192
+[2023-09-02 11:17:24,203][02307] Avg episode reward: 4.692, avg true_objective: 4.192
+[2023-09-02 11:17:49,327][02307] Replay video saved to /content/train_dir/default_experiment/replay.mp4!