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.1700031701.4391a95ca488

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe35b8b0d096d1765b515cb8736640d2c619211ad36fbfa0cb42cbbd34260d3f
+size 2256

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: 3.95 +/- 0.22
+      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 nikxtaco/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/checkpoint_000000002_8192.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebb8b7db46f5efe55b31c40b49344e81c1c7c6bb6b1b7a3b3fbfb8eeb13f528f
+size 34929349

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": 4000,
+  "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=4000",
+  "cli_args": {
+    "env": "doom_health_gathering_supreme",
+    "num_workers": 8,
+    "num_envs_per_worker": 4,
+    "train_for_env_steps": 4000
+  },
+  "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 sha256:01b87e99871d2f8d4d81a63e3acee4121ec1176e9c5d6e8d3e3bf3ea93a90f50
+size 6189816

sf_log.txt

@@ -0,0 +1,484 @@
+[2023-11-15 07:01:43,246][00663] Saving configuration to /content/train_dir/default_experiment/config.json...
+[2023-11-15 07:01:43,251][00663] Rollout worker 0 uses device cpu
+[2023-11-15 07:01:43,255][00663] Rollout worker 1 uses device cpu
+[2023-11-15 07:01:43,256][00663] Rollout worker 2 uses device cpu
+[2023-11-15 07:01:43,258][00663] Rollout worker 3 uses device cpu
+[2023-11-15 07:01:43,262][00663] Rollout worker 4 uses device cpu
+[2023-11-15 07:01:43,264][00663] Rollout worker 5 uses device cpu
+[2023-11-15 07:01:43,265][00663] Rollout worker 6 uses device cpu
+[2023-11-15 07:01:43,267][00663] Rollout worker 7 uses device cpu
+[2023-11-15 07:01:43,416][00663] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-11-15 07:01:43,417][00663] InferenceWorker_p0-w0: min num requests: 2
+[2023-11-15 07:01:43,448][00663] Starting all processes...
+[2023-11-15 07:01:43,450][00663] Starting process learner_proc0
+[2023-11-15 07:01:43,501][00663] Starting all processes...
+[2023-11-15 07:01:43,510][00663] Starting process inference_proc0-0
+[2023-11-15 07:01:43,510][00663] Starting process rollout_proc0
+[2023-11-15 07:01:43,512][00663] Starting process rollout_proc1
+[2023-11-15 07:01:43,513][00663] Starting process rollout_proc2
+[2023-11-15 07:01:43,514][00663] Starting process rollout_proc3
+[2023-11-15 07:01:43,514][00663] Starting process rollout_proc4
+[2023-11-15 07:01:43,514][00663] Starting process rollout_proc5
+[2023-11-15 07:01:43,514][00663] Starting process rollout_proc6
+[2023-11-15 07:01:43,514][00663] Starting process rollout_proc7
+[2023-11-15 07:02:00,834][10761] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-11-15 07:02:00,838][10761] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-11-15 07:02:00,910][10761] Num visible devices: 1
+[2023-11-15 07:02:00,941][10761] Starting seed is not provided
+[2023-11-15 07:02:00,941][10761] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-11-15 07:02:00,941][10761] Initializing actor-critic model on device cuda:0
+[2023-11-15 07:02:00,943][10761] RunningMeanStd input shape: (3, 72, 128)
+[2023-11-15 07:02:00,948][10761] RunningMeanStd input shape: (1,)
+[2023-11-15 07:02:00,987][10780] Worker 1 uses CPU cores [1]
+[2023-11-15 07:02:01,005][10761] ConvEncoder: input_channels=3
+[2023-11-15 07:02:01,011][10779] Worker 0 uses CPU cores [0]
+[2023-11-15 07:02:01,024][10778] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-11-15 07:02:01,027][10778] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-11-15 07:02:01,056][10778] Num visible devices: 1
+[2023-11-15 07:02:01,127][10785] Worker 6 uses CPU cores [0]
+[2023-11-15 07:02:01,251][10786] Worker 7 uses CPU cores [1]
+[2023-11-15 07:02:01,290][10781] Worker 3 uses CPU cores [1]
+[2023-11-15 07:02:01,334][10783] Worker 4 uses CPU cores [0]
+[2023-11-15 07:02:01,373][10782] Worker 2 uses CPU cores [0]
+[2023-11-15 07:02:01,383][10784] Worker 5 uses CPU cores [1]
+[2023-11-15 07:02:01,422][10761] Conv encoder output size: 512
+[2023-11-15 07:02:01,422][10761] Policy head output size: 512
+[2023-11-15 07:02:01,475][10761] Created Actor Critic model with architecture:
+[2023-11-15 07:02:01,475][10761] 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)
+  )
+)
+[2023-11-15 07:02:01,853][10761] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-11-15 07:02:02,202][10761] No checkpoints found
+[2023-11-15 07:02:02,203][10761] Did not load from checkpoint, starting from scratch!
+[2023-11-15 07:02:02,203][10761] Initialized policy 0 weights for model version 0
+[2023-11-15 07:02:02,207][10761] LearnerWorker_p0 finished initialization!
+[2023-11-15 07:02:02,208][10761] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-11-15 07:02:02,385][10778] RunningMeanStd input shape: (3, 72, 128)
+[2023-11-15 07:02:02,386][10778] RunningMeanStd input shape: (1,)
+[2023-11-15 07:02:02,398][10778] ConvEncoder: input_channels=3
+[2023-11-15 07:02:02,495][10778] Conv encoder output size: 512
+[2023-11-15 07:02:02,495][10778] Policy head output size: 512
+[2023-11-15 07:02:02,555][00663] Inference worker 0-0 is ready!
+[2023-11-15 07:02:02,556][00663] All inference workers are ready! Signal rollout workers to start!
+[2023-11-15 07:02:02,758][10783] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:02,759][10782] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:02,760][10779] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:02,761][10785] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:02,770][10784] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:02,763][10781] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:02,772][10780] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:02,773][10786] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:03,403][00663] Heartbeat connected on Batcher_0
+[2023-11-15 07:02:03,409][00663] Heartbeat connected on LearnerWorker_p0
+[2023-11-15 07:02:03,460][00663] Heartbeat connected on InferenceWorker_p0-w0
+[2023-11-15 07:02:03,770][10780] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:03,769][10784] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:04,070][10782] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:04,076][10783] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:04,079][10785] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:04,923][10780] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:04,927][10784] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:04,992][10781] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:05,650][10782] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:05,654][10783] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:05,705][10779] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:06,206][00663] 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)
+[2023-11-15 07:02:06,845][10786] Decorrelating experience for 0 frames...
+[2023-11-15 07:02:07,400][10784] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:07,555][10780] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:07,628][10785] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:07,692][10779] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:08,073][10783] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:08,089][10782] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:09,560][10785] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:09,642][10779] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:09,771][10781] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:09,855][10784] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:09,900][10782] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:10,156][10786] Decorrelating experience for 32 frames...
+[2023-11-15 07:02:10,296][00663] Heartbeat connected on RolloutWorker_w2
+[2023-11-15 07:02:10,303][00663] Heartbeat connected on RolloutWorker_w5
+[2023-11-15 07:02:11,207][00663] 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)
+[2023-11-15 07:02:11,442][10783] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:11,533][10785] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:11,565][10780] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:11,747][00663] Heartbeat connected on RolloutWorker_w4
+[2023-11-15 07:02:11,837][00663] Heartbeat connected on RolloutWorker_w6
+[2023-11-15 07:02:11,969][00663] Heartbeat connected on RolloutWorker_w1
+[2023-11-15 07:02:12,639][10786] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:14,290][10779] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:14,776][10781] Decorrelating experience for 64 frames...
+[2023-11-15 07:02:15,186][00663] Heartbeat connected on RolloutWorker_w0
+[2023-11-15 07:02:16,203][00663] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 120.8. Samples: 1208. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-11-15 07:02:16,206][00663] Avg episode reward: [(0, '2.474')]
+[2023-11-15 07:02:16,576][10786] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:17,569][00663] Heartbeat connected on RolloutWorker_w7
+[2023-11-15 07:02:17,852][10761] Signal inference workers to stop experience collection...
+[2023-11-15 07:02:17,922][10778] InferenceWorker_p0-w0: stopping experience collection
+[2023-11-15 07:02:18,028][10781] Decorrelating experience for 96 frames...
+[2023-11-15 07:02:18,105][00663] Heartbeat connected on RolloutWorker_w3
+[2023-11-15 07:02:18,723][10761] Signal inference workers to resume experience collection...
+[2023-11-15 07:02:18,724][10778] InferenceWorker_p0-w0: resuming experience collection
+[2023-11-15 07:02:20,186][10761] Stopping Batcher_0...
+[2023-11-15 07:02:20,187][10761] Loop batcher_evt_loop terminating...
+[2023-11-15 07:02:20,187][00663] Component Batcher_0 stopped!
+[2023-11-15 07:02:20,196][10761] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000002_8192.pth...
+[2023-11-15 07:02:20,229][00663] Component RolloutWorker_w1 stopped!
+[2023-11-15 07:02:20,228][10780] Stopping RolloutWorker_w1...
+[2023-11-15 07:02:20,234][00663] Component RolloutWorker_w5 stopped!
+[2023-11-15 07:02:20,240][10781] Stopping RolloutWorker_w3...
+[2023-11-15 07:02:20,242][10781] Loop rollout_proc3_evt_loop terminating...
+[2023-11-15 07:02:20,241][00663] Component RolloutWorker_w3 stopped!
+[2023-11-15 07:02:20,234][10784] Stopping RolloutWorker_w5...
+[2023-11-15 07:02:20,235][10780] Loop rollout_proc1_evt_loop terminating...
+[2023-11-15 07:02:20,251][00663] Component RolloutWorker_w7 stopped!
+[2023-11-15 07:02:20,250][10786] Stopping RolloutWorker_w7...
+[2023-11-15 07:02:20,251][10784] Loop rollout_proc5_evt_loop terminating...
+[2023-11-15 07:02:20,254][10786] Loop rollout_proc7_evt_loop terminating...
+[2023-11-15 07:02:20,269][10778] Weights refcount: 2 0
+[2023-11-15 07:02:20,270][10782] Stopping RolloutWorker_w2...
+[2023-11-15 07:02:20,270][00663] Component RolloutWorker_w2 stopped!
+[2023-11-15 07:02:20,272][10778] Stopping InferenceWorker_p0-w0...
+[2023-11-15 07:02:20,273][10778] Loop inference_proc0-0_evt_loop terminating...
+[2023-11-15 07:02:20,273][00663] Component InferenceWorker_p0-w0 stopped!
+[2023-11-15 07:02:20,286][10782] Loop rollout_proc2_evt_loop terminating...
+[2023-11-15 07:02:20,295][00663] Component RolloutWorker_w6 stopped!
+[2023-11-15 07:02:20,303][00663] Component RolloutWorker_w0 stopped!
+[2023-11-15 07:02:20,295][10785] Stopping RolloutWorker_w6...
+[2023-11-15 07:02:20,303][10779] Stopping RolloutWorker_w0...
+[2023-11-15 07:02:20,309][10785] Loop rollout_proc6_evt_loop terminating...
+[2023-11-15 07:02:20,310][10779] Loop rollout_proc0_evt_loop terminating...
+[2023-11-15 07:02:20,321][00663] Component RolloutWorker_w4 stopped!
+[2023-11-15 07:02:20,321][10783] Stopping RolloutWorker_w4...
+[2023-11-15 07:02:20,327][10783] Loop rollout_proc4_evt_loop terminating...
+[2023-11-15 07:02:20,370][10761] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000002_8192.pth...
+[2023-11-15 07:02:20,586][00663] Component LearnerWorker_p0 stopped!
+[2023-11-15 07:02:20,588][00663] Waiting for process learner_proc0 to stop...
+[2023-11-15 07:02:20,586][10761] Stopping LearnerWorker_p0...
+[2023-11-15 07:02:20,591][10761] Loop learner_proc0_evt_loop terminating...
+[2023-11-15 07:02:22,053][00663] Waiting for process inference_proc0-0 to join...
+[2023-11-15 07:02:22,102][00663] Waiting for process rollout_proc0 to join...
+[2023-11-15 07:02:24,502][00663] Waiting for process rollout_proc1 to join...
+[2023-11-15 07:02:24,604][00663] Waiting for process rollout_proc2 to join...
+[2023-11-15 07:02:24,606][00663] Waiting for process rollout_proc3 to join...
+[2023-11-15 07:02:24,609][00663] Waiting for process rollout_proc4 to join...
+[2023-11-15 07:02:24,612][00663] Waiting for process rollout_proc5 to join...
+[2023-11-15 07:02:24,615][00663] Waiting for process rollout_proc6 to join...
+[2023-11-15 07:02:24,620][00663] Waiting for process rollout_proc7 to join...
+[2023-11-15 07:02:24,623][00663] Batcher 0 profile tree view:
+batching: 0.0629, releasing_batches: 0.0004
+[2023-11-15 07:02:24,625][00663] InferenceWorker_p0-w0 profile tree view:
+wait_policy: 0.0000
+  wait_policy_total: 11.1031
+update_model: 0.0531
+  weight_update: 0.0035
+one_step: 0.0117
+  handle_policy_step: 5.3752
+    deserialize: 0.0825, stack: 0.0142, obs_to_device_normalize: 0.7806, forward: 3.7947, send_messages: 0.1665
+    prepare_outputs: 0.4019
+      to_cpu: 0.1972
+[2023-11-15 07:02:24,627][00663] Learner 0 profile tree view:
+misc: 0.0000, prepare_batch: 3.1931
+train: 1.9676
+  epoch_init: 0.0000, minibatch_init: 0.0000, losses_postprocess: 0.0005, kl_divergence: 0.0067, after_optimizer: 0.0663
+  calculate_losses: 0.6155
+    losses_init: 0.0000, forward_head: 0.3481, bptt_initial: 0.1651, tail: 0.0300, advantages_returns: 0.0021, losses: 0.0578
+    bptt: 0.0118
+      bptt_forward_core: 0.0117
+  update: 1.2776
+    clip: 0.0777
+[2023-11-15 07:02:24,629][00663] RolloutWorker_w0 profile tree view:
+wait_for_trajectories: 0.0032, enqueue_policy_requests: 0.6631, env_step: 2.7787, overhead: 0.0796, complete_rollouts: 0.0091
+save_policy_outputs: 0.0847
+  split_output_tensors: 0.0449
+[2023-11-15 07:02:24,632][00663] RolloutWorker_w7 profile tree view:
+wait_for_trajectories: 0.0005, enqueue_policy_requests: 0.1984, env_step: 1.3664, overhead: 0.0326, complete_rollouts: 0.0003
+save_policy_outputs: 0.0129
+  split_output_tensors: 0.0069
+[2023-11-15 07:02:24,636][00663] Loop Runner_EvtLoop terminating...
+[2023-11-15 07:02:24,638][00663] Runner profile tree view:
+main_loop: 41.1905
+[2023-11-15 07:02:24,640][00663] Collected {0: 8192}, FPS: 198.9
+[2023-11-15 07:02:45,729][00663] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2023-11-15 07:02:45,730][00663] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-11-15 07:02:45,734][00663] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-11-15 07:02:45,739][00663] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-11-15 07:02:45,742][00663] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-11-15 07:02:45,743][00663] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-11-15 07:02:45,749][00663] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file!
+[2023-11-15 07:02:45,750][00663] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-11-15 07:02:45,751][00663] Adding new argument 'push_to_hub'=False that is not in the saved config file!
+[2023-11-15 07:02:45,753][00663] Adding new argument 'hf_repository'=None that is not in the saved config file!
+[2023-11-15 07:02:45,755][00663] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-11-15 07:02:45,756][00663] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-11-15 07:02:45,757][00663] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-11-15 07:02:45,758][00663] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-11-15 07:02:45,760][00663] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-11-15 07:02:45,797][00663] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-11-15 07:02:45,801][00663] RunningMeanStd input shape: (3, 72, 128)
+[2023-11-15 07:02:45,806][00663] RunningMeanStd input shape: (1,)
+[2023-11-15 07:02:45,821][00663] ConvEncoder: input_channels=3
+[2023-11-15 07:02:45,925][00663] Conv encoder output size: 512
+[2023-11-15 07:02:45,926][00663] Policy head output size: 512
+[2023-11-15 07:02:53,558][00663] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000002_8192.pth...
+[2023-11-15 07:02:57,671][00663] Num frames 100...
+[2023-11-15 07:02:57,861][00663] Num frames 200...
+[2023-11-15 07:02:58,054][00663] Num frames 300...
+[2023-11-15 07:02:58,250][00663] Num frames 400...
+[2023-11-15 07:02:58,398][00663] Avg episode rewards: #0: 5.480, true rewards: #0: 4.480
+[2023-11-15 07:02:58,401][00663] Avg episode reward: 5.480, avg true_objective: 4.480
+[2023-11-15 07:02:58,505][00663] Num frames 500...
+[2023-11-15 07:02:58,692][00663] Num frames 600...
+[2023-11-15 07:02:58,874][00663] Num frames 700...
+[2023-11-15 07:02:59,057][00663] Num frames 800...
+[2023-11-15 07:02:59,155][00663] Avg episode rewards: #0: 4.660, true rewards: #0: 4.160
+[2023-11-15 07:02:59,156][00663] Avg episode reward: 4.660, avg true_objective: 4.160
+[2023-11-15 07:02:59,251][00663] Num frames 900...
+[2023-11-15 07:02:59,388][00663] Num frames 1000...
+[2023-11-15 07:02:59,534][00663] Num frames 1100...
+[2023-11-15 07:02:59,667][00663] Num frames 1200...
+[2023-11-15 07:02:59,826][00663] Avg episode rewards: #0: 4.933, true rewards: #0: 4.267
+[2023-11-15 07:02:59,828][00663] Avg episode reward: 4.933, avg true_objective: 4.267
+[2023-11-15 07:02:59,858][00663] Num frames 1300...
+[2023-11-15 07:02:59,983][00663] Num frames 1400...
+[2023-11-15 07:03:00,112][00663] Num frames 1500...
+[2023-11-15 07:03:00,241][00663] Num frames 1600...
+[2023-11-15 07:03:00,367][00663] Num frames 1700...
+[2023-11-15 07:03:00,462][00663] Avg episode rewards: #0: 5.070, true rewards: #0: 4.320
+[2023-11-15 07:03:00,463][00663] Avg episode reward: 5.070, avg true_objective: 4.320
+[2023-11-15 07:03:00,561][00663] Num frames 1800...
+[2023-11-15 07:03:00,688][00663] Num frames 1900...
+[2023-11-15 07:03:00,815][00663] Num frames 2000...
+[2023-11-15 07:03:00,939][00663] Num frames 2100...
+[2023-11-15 07:03:01,047][00663] Avg episode rewards: #0: 4.888, true rewards: #0: 4.288
+[2023-11-15 07:03:01,050][00663] Avg episode reward: 4.888, avg true_objective: 4.288
+[2023-11-15 07:03:01,123][00663] Num frames 2200...
+[2023-11-15 07:03:01,253][00663] Num frames 2300...
+[2023-11-15 07:03:01,386][00663] Num frames 2400...
+[2023-11-15 07:03:01,514][00663] Num frames 2500...
+[2023-11-15 07:03:01,698][00663] Avg episode rewards: #0: 4.987, true rewards: #0: 4.320
+[2023-11-15 07:03:01,700][00663] Avg episode reward: 4.987, avg true_objective: 4.320
+[2023-11-15 07:03:01,718][00663] Num frames 2600...
+[2023-11-15 07:03:01,862][00663] Num frames 2700...
+[2023-11-15 07:03:01,996][00663] Num frames 2800...
+[2023-11-15 07:03:02,124][00663] Num frames 2900...
+[2023-11-15 07:03:02,278][00663] Avg episode rewards: #0: 4.823, true rewards: #0: 4.251
+[2023-11-15 07:03:02,279][00663] Avg episode reward: 4.823, avg true_objective: 4.251
+[2023-11-15 07:03:02,313][00663] Num frames 3000...
+[2023-11-15 07:03:02,439][00663] Num frames 3100...
+[2023-11-15 07:03:02,585][00663] Num frames 3200...
+[2023-11-15 07:03:02,713][00663] Num frames 3300...
+[2023-11-15 07:03:02,845][00663] Avg episode rewards: #0: 4.700, true rewards: #0: 4.200
+[2023-11-15 07:03:02,846][00663] Avg episode reward: 4.700, avg true_objective: 4.200
+[2023-11-15 07:03:02,903][00663] Num frames 3400...
+[2023-11-15 07:03:03,028][00663] Num frames 3500...
+[2023-11-15 07:03:03,164][00663] Num frames 3600...
+[2023-11-15 07:03:03,291][00663] Num frames 3700...
+[2023-11-15 07:03:03,401][00663] Avg episode rewards: #0: 4.604, true rewards: #0: 4.160
+[2023-11-15 07:03:03,402][00663] Avg episode reward: 4.604, avg true_objective: 4.160
+[2023-11-15 07:03:03,475][00663] Num frames 3800...
+[2023-11-15 07:03:03,606][00663] Num frames 3900...
+[2023-11-15 07:03:03,733][00663] Num frames 4000...
+[2023-11-15 07:03:03,860][00663] Num frames 4100...
+[2023-11-15 07:03:03,954][00663] Avg episode rewards: #0: 4.528, true rewards: #0: 4.128
+[2023-11-15 07:03:03,955][00663] Avg episode reward: 4.528, avg true_objective: 4.128
+[2023-11-15 07:03:30,284][00663] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
+[2023-11-15 07:03:44,458][00663] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2023-11-15 07:03:44,463][00663] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-11-15 07:03:44,465][00663] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-11-15 07:03:44,467][00663] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-11-15 07:03:44,473][00663] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-11-15 07:03:44,474][00663] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-11-15 07:03:44,476][00663] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
+[2023-11-15 07:03:44,477][00663] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-11-15 07:03:44,478][00663] Adding new argument 'push_to_hub'=True that is not in the saved config file!
+[2023-11-15 07:03:44,484][00663] Adding new argument 'hf_repository'='nikxtaco/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
+[2023-11-15 07:03:44,485][00663] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-11-15 07:03:44,486][00663] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-11-15 07:03:44,487][00663] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-11-15 07:03:44,488][00663] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-11-15 07:03:44,489][00663] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-11-15 07:03:44,548][00663] RunningMeanStd input shape: (3, 72, 128)
+[2023-11-15 07:03:44,553][00663] RunningMeanStd input shape: (1,)
+[2023-11-15 07:03:44,575][00663] ConvEncoder: input_channels=3
+[2023-11-15 07:03:44,652][00663] Conv encoder output size: 512
+[2023-11-15 07:03:44,655][00663] Policy head output size: 512
+[2023-11-15 07:03:44,682][00663] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000002_8192.pth...
+[2023-11-15 07:03:45,285][00663] Num frames 100...
+[2023-11-15 07:03:45,421][00663] Num frames 200...
+[2023-11-15 07:03:45,563][00663] Num frames 300...
+[2023-11-15 07:03:45,734][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:45,737][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:45,763][00663] Num frames 400...
+[2023-11-15 07:03:45,894][00663] Num frames 500...
+[2023-11-15 07:03:46,058][00663] Num frames 600...
+[2023-11-15 07:03:46,234][00663] Num frames 700...
+[2023-11-15 07:03:46,379][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:46,380][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:46,428][00663] Num frames 800...
+[2023-11-15 07:03:46,552][00663] Num frames 900...
+[2023-11-15 07:03:46,685][00663] Num frames 1000...
+[2023-11-15 07:03:46,815][00663] Num frames 1100...
+[2023-11-15 07:03:46,939][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:46,942][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:47,005][00663] Num frames 1200...
+[2023-11-15 07:03:47,138][00663] Num frames 1300...
+[2023-11-15 07:03:47,265][00663] Num frames 1400...
+[2023-11-15 07:03:47,393][00663] Num frames 1500...
+[2023-11-15 07:03:47,498][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:47,500][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:47,584][00663] Num frames 1600...
+[2023-11-15 07:03:47,719][00663] Num frames 1700...
+[2023-11-15 07:03:47,849][00663] Num frames 1800...
+[2023-11-15 07:03:47,973][00663] Num frames 1900...
+[2023-11-15 07:03:48,055][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:48,057][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:48,163][00663] Num frames 2000...
+[2023-11-15 07:03:48,289][00663] Num frames 2100...
+[2023-11-15 07:03:48,419][00663] Num frames 2200...
+[2023-11-15 07:03:48,545][00663] Num frames 2300...
+[2023-11-15 07:03:48,606][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:48,607][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:48,741][00663] Num frames 2400...
+[2023-11-15 07:03:48,869][00663] Num frames 2500...
+[2023-11-15 07:03:48,992][00663] Num frames 2600...
+[2023-11-15 07:03:49,160][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:49,161][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:49,181][00663] Num frames 2700...
+[2023-11-15 07:03:49,313][00663] Num frames 2800...
+[2023-11-15 07:03:49,439][00663] Num frames 2900...
+[2023-11-15 07:03:49,564][00663] Num frames 3000...
+[2023-11-15 07:03:49,711][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:49,713][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:49,757][00663] Num frames 3100...
+[2023-11-15 07:03:49,883][00663] Num frames 3200...
+[2023-11-15 07:03:50,006][00663] Num frames 3300...
+[2023-11-15 07:03:50,148][00663] Num frames 3400...
+[2023-11-15 07:03:50,274][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:50,276][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:03:50,335][00663] Num frames 3500...
+[2023-11-15 07:03:50,460][00663] Num frames 3600...
+[2023-11-15 07:03:50,584][00663] Num frames 3700...
+[2023-11-15 07:03:50,718][00663] Num frames 3800...
+[2023-11-15 07:03:50,834][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:03:50,836][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:04:09,945][00663] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
+[2023-11-15 07:04:24,386][00663] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2023-11-15 07:04:24,389][00663] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-11-15 07:04:24,391][00663] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-11-15 07:04:24,393][00663] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-11-15 07:04:24,396][00663] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-11-15 07:04:24,398][00663] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-11-15 07:04:24,400][00663] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
+[2023-11-15 07:04:24,402][00663] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-11-15 07:04:24,403][00663] Adding new argument 'push_to_hub'=True that is not in the saved config file!
+[2023-11-15 07:04:24,404][00663] Adding new argument 'hf_repository'='nikxtaco/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
+[2023-11-15 07:04:24,405][00663] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-11-15 07:04:24,406][00663] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-11-15 07:04:24,407][00663] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-11-15 07:04:24,409][00663] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-11-15 07:04:24,410][00663] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-11-15 07:04:24,447][00663] RunningMeanStd input shape: (3, 72, 128)
+[2023-11-15 07:04:24,449][00663] RunningMeanStd input shape: (1,)
+[2023-11-15 07:04:24,462][00663] ConvEncoder: input_channels=3
+[2023-11-15 07:04:24,498][00663] Conv encoder output size: 512
+[2023-11-15 07:04:24,499][00663] Policy head output size: 512
+[2023-11-15 07:04:24,518][00663] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000002_8192.pth...
+[2023-11-15 07:04:24,954][00663] Num frames 100...
+[2023-11-15 07:04:25,083][00663] Num frames 200...
+[2023-11-15 07:04:25,208][00663] Num frames 300...
+[2023-11-15 07:04:25,402][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:04:25,405][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:04:25,438][00663] Num frames 400...
+[2023-11-15 07:04:25,627][00663] Num frames 500...
+[2023-11-15 07:04:25,826][00663] Num frames 600...
+[2023-11-15 07:04:26,017][00663] Num frames 700...
+[2023-11-15 07:04:26,206][00663] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2023-11-15 07:04:26,208][00663] Avg episode reward: 3.840, avg true_objective: 3.840
+[2023-11-15 07:04:26,273][00663] Num frames 800...
+[2023-11-15 07:04:26,472][00663] Num frames 900...
+[2023-11-15 07:04:26,666][00663] Num frames 1000...
+[2023-11-15 07:04:26,875][00663] Num frames 1100...
+[2023-11-15 07:04:27,057][00663] Num frames 1200...
+[2023-11-15 07:04:27,142][00663] Avg episode rewards: #0: 4.387, true rewards: #0: 4.053
+[2023-11-15 07:04:27,145][00663] Avg episode reward: 4.387, avg true_objective: 4.053
+[2023-11-15 07:04:27,312][00663] Num frames 1300...
+[2023-11-15 07:04:27,525][00663] Num frames 1400...
+[2023-11-15 07:04:27,728][00663] Num frames 1500...
+[2023-11-15 07:04:27,956][00663] Num frames 1600...
+[2023-11-15 07:04:28,012][00663] Avg episode rewards: #0: 4.250, true rewards: #0: 4.000
+[2023-11-15 07:04:28,014][00663] Avg episode reward: 4.250, avg true_objective: 4.000
+[2023-11-15 07:04:28,206][00663] Num frames 1700...
+[2023-11-15 07:04:28,392][00663] Num frames 1800...
+[2023-11-15 07:04:28,570][00663] Num frames 1900...
+[2023-11-15 07:04:28,780][00663] Avg episode rewards: #0: 4.168, true rewards: #0: 3.968
+[2023-11-15 07:04:28,782][00663] Avg episode reward: 4.168, avg true_objective: 3.968
+[2023-11-15 07:04:28,815][00663] Num frames 2000...
+[2023-11-15 07:04:28,992][00663] Num frames 2100...
+[2023-11-15 07:04:29,175][00663] Num frames 2200...
+[2023-11-15 07:04:29,362][00663] Num frames 2300...
+[2023-11-15 07:04:29,549][00663] Avg episode rewards: #0: 4.113, true rewards: #0: 3.947
+[2023-11-15 07:04:29,551][00663] Avg episode reward: 4.113, avg true_objective: 3.947
+[2023-11-15 07:04:29,615][00663] Num frames 2400...
+[2023-11-15 07:04:29,807][00663] Num frames 2500...
+[2023-11-15 07:04:29,995][00663] Num frames 2600...
+[2023-11-15 07:04:30,176][00663] Num frames 2700...
+[2023-11-15 07:04:30,332][00663] Avg episode rewards: #0: 4.074, true rewards: #0: 3.931
+[2023-11-15 07:04:30,335][00663] Avg episode reward: 4.074, avg true_objective: 3.931
+[2023-11-15 07:04:30,431][00663] Num frames 2800...
+[2023-11-15 07:04:30,613][00663] Num frames 2900...
+[2023-11-15 07:04:30,802][00663] Num frames 3000...
+[2023-11-15 07:04:30,988][00663] Num frames 3100...
+[2023-11-15 07:04:31,163][00663] Avg episode rewards: #0: 4.353, true rewards: #0: 3.977
+[2023-11-15 07:04:31,164][00663] Avg episode reward: 4.353, avg true_objective: 3.977
+[2023-11-15 07:04:31,190][00663] Num frames 3200...
+[2023-11-15 07:04:31,325][00663] Num frames 3300...
+[2023-11-15 07:04:31,455][00663] Num frames 3400...
+[2023-11-15 07:04:31,583][00663] Num frames 3500...
+[2023-11-15 07:04:31,725][00663] Avg episode rewards: #0: 4.296, true rewards: #0: 3.962
+[2023-11-15 07:04:31,726][00663] Avg episode reward: 4.296, avg true_objective: 3.962
+[2023-11-15 07:04:31,772][00663] Num frames 3600...
+[2023-11-15 07:04:31,903][00663] Num frames 3700...
+[2023-11-15 07:04:32,033][00663] Num frames 3800...
+[2023-11-15 07:04:32,159][00663] Num frames 3900...
+[2023-11-15 07:04:32,287][00663] Avg episode rewards: #0: 4.250, true rewards: #0: 3.950
+[2023-11-15 07:04:32,288][00663] Avg episode reward: 4.250, avg true_objective: 3.950
+[2023-11-15 07:04:53,857][00663] Replay video saved to /content/train_dir/default_experiment/replay.mp4!