Upload . with huggingface_hub

.summary/0/events.out.tfevents.1677403925.14111cc73324

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b064f2402f25d91bf4706a638e4917af6f13be9e5f4b9eff531bc3c5c376d589
+size 2086

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: 4.10 +/- 0.81
+      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 chavicoski/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=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=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_000000004_16384.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2784c5c0f2fa7f0f8f5596c27c93bda6dc515070626fad4b0cd067093ce1877
+size 34928836

config.json

@@ -0,0 +1,142 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "doom_health_gathering_supreme",
+  "experiment": "default_experiment",
+  "train_dir": "/workspace/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": 12,
+  "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": 10000,
+  "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=12 --num_envs_per_worker=4 --train_for_env_steps=10000",
+  "cli_args": {
+    "env": "doom_health_gathering_supreme",
+    "num_workers": 12,
+    "num_envs_per_worker": 4,
+    "train_for_env_steps": 10000
+  },
+  "git_hash": "unknown",
+  "git_repo_name": "not a git repository"
+}

sf_log.txt

@@ -0,0 +1,356 @@
+[2023-02-26 09:32:07,567][00001] Saving configuration to /workspace/train_dir/default_experiment/config.json...
+[2023-02-26 09:32:07,568][00001] Rollout worker 0 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 1 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 2 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 3 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 4 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 5 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 6 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 7 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 8 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 9 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 10 uses device cpu
+[2023-02-26 09:32:07,568][00001] Rollout worker 11 uses device cpu
+[2023-02-26 09:32:07,624][00001] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-02-26 09:32:07,624][00001] InferenceWorker_p0-w0: min num requests: 4
+[2023-02-26 09:32:07,647][00001] Starting all processes...
+[2023-02-26 09:32:07,647][00001] Starting process learner_proc0
+[2023-02-26 09:32:08,374][00001] Starting all processes...
+[2023-02-26 09:32:08,377][00001] Starting process inference_proc0-0
+[2023-02-26 09:32:08,377][00001] Starting process rollout_proc0
+[2023-02-26 09:32:08,377][00001] Starting process rollout_proc1
+[2023-02-26 09:32:08,378][00141] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-02-26 09:32:08,378][00141] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-02-26 09:32:08,377][00001] Starting process rollout_proc2
+[2023-02-26 09:32:08,377][00001] Starting process rollout_proc3
+[2023-02-26 09:32:08,377][00001] Starting process rollout_proc4
+[2023-02-26 09:32:08,378][00001] Starting process rollout_proc5
+[2023-02-26 09:32:08,378][00001] Starting process rollout_proc6
+[2023-02-26 09:32:08,387][00141] Num visible devices: 1
+[2023-02-26 09:32:08,378][00001] Starting process rollout_proc7
+[2023-02-26 09:32:08,379][00001] Starting process rollout_proc8
+[2023-02-26 09:32:08,380][00001] Starting process rollout_proc9
+[2023-02-26 09:32:08,381][00001] Starting process rollout_proc10
+[2023-02-26 09:32:08,383][00001] Starting process rollout_proc11
+[2023-02-26 09:32:08,422][00141] Starting seed is not provided
+[2023-02-26 09:32:08,422][00141] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-02-26 09:32:08,422][00141] Initializing actor-critic model on device cuda:0
+[2023-02-26 09:32:08,422][00141] RunningMeanStd input shape: (3, 72, 128)
+[2023-02-26 09:32:08,423][00141] RunningMeanStd input shape: (1,)
+[2023-02-26 09:32:08,438][00141] ConvEncoder: input_channels=3
+[2023-02-26 09:32:08,565][00141] Conv encoder output size: 512
+[2023-02-26 09:32:08,566][00141] Policy head output size: 512
+[2023-02-26 09:32:08,579][00141] Created Actor Critic model with architecture:
+[2023-02-26 09:32:08,579][00141] 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)
+  )
+)
+[2023-02-26 09:32:09,423][00201] Worker 10 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,462][00197] Worker 8 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,464][00195] Worker 4 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,486][00190] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-02-26 09:32:09,486][00192] Worker 2 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,486][00190] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-02-26 09:32:09,488][00196] Worker 6 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,493][00189] Worker 1 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,497][00190] Num visible devices: 1
+[2023-02-26 09:32:09,507][00191] Worker 0 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,513][00200] Worker 9 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,523][00194] Worker 5 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,534][00199] Worker 11 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,542][00198] Worker 7 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:09,561][00193] Worker 3 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]
+[2023-02-26 09:32:10,323][00141] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-02-26 09:32:10,324][00141] No checkpoints found
+[2023-02-26 09:32:10,324][00141] Did not load from checkpoint, starting from scratch!
+[2023-02-26 09:32:10,324][00141] Initialized policy 0 weights for model version 0
+[2023-02-26 09:32:10,325][00141] LearnerWorker_p0 finished initialization!
+[2023-02-26 09:32:10,325][00141] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-02-26 09:32:10,383][00190] RunningMeanStd input shape: (3, 72, 128)
+[2023-02-26 09:32:10,383][00190] RunningMeanStd input shape: (1,)
+[2023-02-26 09:32:10,391][00190] ConvEncoder: input_channels=3
+[2023-02-26 09:32:10,454][00190] Conv encoder output size: 512
+[2023-02-26 09:32:10,454][00190] Policy head output size: 512
+[2023-02-26 09:32:10,996][00001] 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)
+[2023-02-26 09:32:11,166][00001] Inference worker 0-0 is ready!
+[2023-02-26 09:32:11,166][00001] All inference workers are ready! Signal rollout workers to start!
+[2023-02-26 09:32:11,194][00196] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,199][00199] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,206][00200] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,206][00197] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,214][00193] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,214][00201] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,220][00198] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,227][00191] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,233][00189] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,234][00192] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,235][00194] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,235][00195] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:11,359][00196] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,359][00199] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,401][00197] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,401][00193] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,401][00200] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,407][00191] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,407][00192] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,534][00201] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,578][00197] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,579][00193] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,582][00198] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,582][00194] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,586][00199] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,586][00191] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,586][00192] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,673][00201] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,691][00189] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,719][00194] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,768][00200] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,772][00197] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:11,776][00193] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:11,778][00196] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,778][00199] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:11,828][00198] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,860][00191] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:11,884][00194] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:11,943][00200] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:11,955][00196] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:11,962][00197] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:11,964][00195] Decorrelating experience for 0 frames...
+[2023-02-26 09:32:11,968][00193] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:11,982][00189] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:11,992][00198] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:12,096][00199] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,107][00192] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:12,133][00191] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,140][00201] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:12,156][00198] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,157][00194] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,281][00196] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,305][00192] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,318][00195] Decorrelating experience for 32 frames...
+[2023-02-26 09:32:12,321][00200] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,489][00201] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,502][00189] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:12,511][00195] Decorrelating experience for 64 frames...
+[2023-02-26 09:32:12,629][00141] Signal inference workers to stop experience collection...
+[2023-02-26 09:32:12,632][00190] InferenceWorker_p0-w0: stopping experience collection
+[2023-02-26 09:32:12,696][00189] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:12,698][00195] Decorrelating experience for 96 frames...
+[2023-02-26 09:32:13,348][00141] Signal inference workers to resume experience collection...
+[2023-02-26 09:32:13,348][00190] InferenceWorker_p0-w0: resuming experience collection
+[2023-02-26 09:32:14,002][00141] Stopping Batcher_0...
+[2023-02-26 09:32:14,002][00001] Component Batcher_0 stopped!
+[2023-02-26 09:32:14,002][00141] Saving /workspace/train_dir/default_experiment/checkpoint_p0/checkpoint_000000004_16384.pth...
+[2023-02-26 09:32:14,010][00198] Stopping RolloutWorker_w7...
+[2023-02-26 09:32:14,010][00001] Component RolloutWorker_w7 stopped!
+[2023-02-26 09:32:14,011][00198] Loop rollout_proc7_evt_loop terminating...
+[2023-02-26 09:32:14,011][00001] Component RolloutWorker_w1 stopped!
+[2023-02-26 09:32:14,002][00141] Loop batcher_evt_loop terminating...
+[2023-02-26 09:32:14,011][00189] Stopping RolloutWorker_w1...
+[2023-02-26 09:32:14,011][00001] Component RolloutWorker_w10 stopped!
+[2023-02-26 09:32:14,011][00201] Stopping RolloutWorker_w10...
+[2023-02-26 09:32:14,011][00195] Stopping RolloutWorker_w4...
+[2023-02-26 09:32:14,011][00001] Component RolloutWorker_w4 stopped!
+[2023-02-26 09:32:14,011][00201] Loop rollout_proc10_evt_loop terminating...
+[2023-02-26 09:32:14,011][00197] Stopping RolloutWorker_w8...
+[2023-02-26 09:32:14,011][00189] Loop rollout_proc1_evt_loop terminating...
+[2023-02-26 09:32:14,011][00195] Loop rollout_proc4_evt_loop terminating...
+[2023-02-26 09:32:14,011][00001] Component RolloutWorker_w8 stopped!
+[2023-02-26 09:32:14,011][00199] Stopping RolloutWorker_w11...
+[2023-02-26 09:32:14,011][00001] Component RolloutWorker_w11 stopped!
+[2023-02-26 09:32:14,011][00197] Loop rollout_proc8_evt_loop terminating...
+[2023-02-26 09:32:14,011][00001] Component RolloutWorker_w2 stopped!
+[2023-02-26 09:32:14,011][00191] Stopping RolloutWorker_w0...
+[2023-02-26 09:32:14,011][00200] Stopping RolloutWorker_w9...
+[2023-02-26 09:32:14,011][00192] Stopping RolloutWorker_w2...
+[2023-02-26 09:32:14,011][00193] Stopping RolloutWorker_w3...
+[2023-02-26 09:32:14,011][00199] Loop rollout_proc11_evt_loop terminating...
+[2023-02-26 09:32:14,011][00196] Stopping RolloutWorker_w6...
+[2023-02-26 09:32:14,012][00001] Component RolloutWorker_w9 stopped!
+[2023-02-26 09:32:14,012][00191] Loop rollout_proc0_evt_loop terminating...
+[2023-02-26 09:32:14,012][00001] Component RolloutWorker_w3 stopped!
+[2023-02-26 09:32:14,011][00194] Stopping RolloutWorker_w5...
+[2023-02-26 09:32:14,012][00200] Loop rollout_proc9_evt_loop terminating...
+[2023-02-26 09:32:14,012][00193] Loop rollout_proc3_evt_loop terminating...
+[2023-02-26 09:32:14,012][00001] Component RolloutWorker_w0 stopped!
+[2023-02-26 09:32:14,012][00196] Loop rollout_proc6_evt_loop terminating...
+[2023-02-26 09:32:14,012][00192] Loop rollout_proc2_evt_loop terminating...
+[2023-02-26 09:32:14,012][00001] Component RolloutWorker_w6 stopped!
+[2023-02-26 09:32:14,012][00194] Loop rollout_proc5_evt_loop terminating...
+[2023-02-26 09:32:14,012][00001] Component RolloutWorker_w5 stopped!
+[2023-02-26 09:32:14,018][00190] Weights refcount: 2 0
+[2023-02-26 09:32:14,020][00001] Component InferenceWorker_p0-w0 stopped!
+[2023-02-26 09:32:14,020][00190] Stopping InferenceWorker_p0-w0...
+[2023-02-26 09:32:14,021][00190] Loop inference_proc0-0_evt_loop terminating...
+[2023-02-26 09:32:14,053][00141] Saving /workspace/train_dir/default_experiment/checkpoint_p0/checkpoint_000000004_16384.pth...
+[2023-02-26 09:32:14,118][00141] Stopping LearnerWorker_p0...
+[2023-02-26 09:32:14,118][00001] Component LearnerWorker_p0 stopped!
+[2023-02-26 09:32:14,119][00141] Loop learner_proc0_evt_loop terminating...
+[2023-02-26 09:32:14,119][00001] Waiting for process learner_proc0 to stop...
+[2023-02-26 09:32:14,900][00001] Waiting for process inference_proc0-0 to join...
+[2023-02-26 09:32:14,901][00001] Waiting for process rollout_proc0 to join...
+[2023-02-26 09:32:14,901][00001] Waiting for process rollout_proc1 to join...
+[2023-02-26 09:32:14,901][00001] Waiting for process rollout_proc2 to join...
+[2023-02-26 09:32:14,901][00001] Waiting for process rollout_proc3 to join...
+[2023-02-26 09:32:14,902][00001] Waiting for process rollout_proc4 to join...
+[2023-02-26 09:32:14,902][00001] Waiting for process rollout_proc5 to join...
+[2023-02-26 09:32:14,902][00001] Waiting for process rollout_proc6 to join...
+[2023-02-26 09:32:14,902][00001] Waiting for process rollout_proc7 to join...
+[2023-02-26 09:32:14,903][00001] Waiting for process rollout_proc8 to join...
+[2023-02-26 09:32:14,903][00001] Waiting for process rollout_proc9 to join...
+[2023-02-26 09:32:14,903][00001] Waiting for process rollout_proc10 to join...
+[2023-02-26 09:32:14,903][00001] Waiting for process rollout_proc11 to join...
+[2023-02-26 09:32:14,904][00001] Batcher 0 profile tree view:
+batching: 0.0462, releasing_batches: 0.0008
+[2023-02-26 09:32:14,904][00001] InferenceWorker_p0-w0 profile tree view:
+wait_policy: 0.0000
+  wait_policy_total: 0.8600
+update_model: 0.2093
+  weight_update: 0.0513
+one_step: 0.0016
+  handle_policy_step: 0.7327
+    deserialize: 0.0239, stack: 0.0026, obs_to_device_normalize: 0.1050, forward: 0.4757, send_messages: 0.0396
+    prepare_outputs: 0.0622
+      to_cpu: 0.0383
+[2023-02-26 09:32:14,904][00001] Learner 0 profile tree view:
+misc: 0.0000, prepare_batch: 1.1570
+train: 0.2483
+  epoch_init: 0.0000, minibatch_init: 0.0000, losses_postprocess: 0.0007, kl_divergence: 0.0010, after_optimizer: 0.0080
+  calculate_losses: 0.0434
+    losses_init: 0.0000, forward_head: 0.0259, bptt_initial: 0.0108, tail: 0.0013, advantages_returns: 0.0005, losses: 0.0024
+    bptt: 0.0021
+      bptt_forward_core: 0.0020
+  update: 0.1943
+    clip: 0.0026
+[2023-02-26 09:32:14,904][00001] RolloutWorker_w0 profile tree view:
+wait_for_trajectories: 0.0006, enqueue_policy_requests: 0.0198, env_step: 0.3360, overhead: 0.0196, complete_rollouts: 0.0005
+save_policy_outputs: 0.0217
+  split_output_tensors: 0.0106
+[2023-02-26 09:32:14,904][00001] RolloutWorker_w11 profile tree view:
+wait_for_trajectories: 0.0006, enqueue_policy_requests: 0.0212, env_step: 0.3282, overhead: 0.0214, complete_rollouts: 0.0006
+save_policy_outputs: 0.0236
+  split_output_tensors: 0.0113
+[2023-02-26 09:32:14,905][00001] Loop Runner_EvtLoop terminating...
+[2023-02-26 09:32:14,905][00001] Runner profile tree view:
+main_loop: 7.2583
+[2023-02-26 09:32:14,905][00001] Collected {0: 16384}, FPS: 2257.3
+[2023-02-26 09:32:14,921][00001] Loading existing experiment configuration from /workspace/train_dir/default_experiment/config.json
+[2023-02-26 09:32:14,922][00001] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-02-26 09:32:14,922][00001] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-02-26 09:32:14,922][00001] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-02-26 09:32:14,922][00001] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-02-26 09:32:14,922][00001] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-02-26 09:32:14,922][00001] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
+[2023-02-26 09:32:14,922][00001] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-02-26 09:32:14,922][00001] Adding new argument 'push_to_hub'=True that is not in the saved config file!
+[2023-02-26 09:32:14,923][00001] Adding new argument 'hf_repository'='chavicoski/vizdoom_health_gathering_supreme' that is not in the saved config file!
+[2023-02-26 09:32:14,923][00001] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-02-26 09:32:14,923][00001] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-02-26 09:32:14,923][00001] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-02-26 09:32:14,923][00001] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-02-26 09:32:14,923][00001] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-02-26 09:32:14,930][00001] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-02-26 09:32:14,930][00001] RunningMeanStd input shape: (3, 72, 128)
+[2023-02-26 09:32:14,931][00001] RunningMeanStd input shape: (1,)
+[2023-02-26 09:32:14,945][00001] ConvEncoder: input_channels=3
+[2023-02-26 09:32:15,033][00001] Conv encoder output size: 512
+[2023-02-26 09:32:15,034][00001] Policy head output size: 512
+[2023-02-26 09:32:16,298][00001] Loading state from checkpoint /workspace/train_dir/default_experiment/checkpoint_p0/checkpoint_000000004_16384.pth...
+[2023-02-26 09:32:16,922][00001] Num frames 100...
+[2023-02-26 09:32:17,014][00001] Num frames 200...
+[2023-02-26 09:32:17,108][00001] Num frames 300...
+[2023-02-26 09:32:17,200][00001] Num frames 400...
+[2023-02-26 09:32:17,293][00001] Num frames 500...
+[2023-02-26 09:32:17,386][00001] Avg episode rewards: #0: 7.440, true rewards: #0: 5.440
+[2023-02-26 09:32:17,387][00001] Avg episode reward: 7.440, avg true_objective: 5.440
+[2023-02-26 09:32:17,463][00001] Num frames 600...
+[2023-02-26 09:32:17,556][00001] Num frames 700...
+[2023-02-26 09:32:17,649][00001] Num frames 800...
+[2023-02-26 09:32:17,743][00001] Num frames 900...
+[2023-02-26 09:32:17,837][00001] Num frames 1000...
+[2023-02-26 09:32:17,972][00001] Avg episode rewards: #0: 7.940, true rewards: #0: 5.440
+[2023-02-26 09:32:17,972][00001] Avg episode reward: 7.940, avg true_objective: 5.440
+[2023-02-26 09:32:17,988][00001] Num frames 1100...
+[2023-02-26 09:32:18,095][00001] Num frames 1200...
+[2023-02-26 09:32:18,188][00001] Num frames 1300...
+[2023-02-26 09:32:18,281][00001] Num frames 1400...
+[2023-02-26 09:32:18,404][00001] Avg episode rewards: #0: 6.573, true rewards: #0: 4.907
+[2023-02-26 09:32:18,404][00001] Avg episode reward: 6.573, avg true_objective: 4.907
+[2023-02-26 09:32:18,442][00001] Num frames 1500...
+[2023-02-26 09:32:18,543][00001] Num frames 1600...
+[2023-02-26 09:32:18,635][00001] Num frames 1700...
+[2023-02-26 09:32:18,728][00001] Num frames 1800...
+[2023-02-26 09:32:18,832][00001] Avg episode rewards: #0: 5.890, true rewards: #0: 4.640
+[2023-02-26 09:32:18,833][00001] Avg episode reward: 5.890, avg true_objective: 4.640
+[2023-02-26 09:32:18,890][00001] Num frames 1900...
+[2023-02-26 09:32:18,986][00001] Num frames 2000...
+[2023-02-26 09:32:19,079][00001] Num frames 2100...
+[2023-02-26 09:32:19,173][00001] Num frames 2200...
+[2023-02-26 09:32:19,267][00001] Num frames 2300...
+[2023-02-26 09:32:19,324][00001] Avg episode rewards: #0: 5.808, true rewards: #0: 4.608
+[2023-02-26 09:32:19,324][00001] Avg episode reward: 5.808, avg true_objective: 4.608
+[2023-02-26 09:32:19,440][00001] Num frames 2400...
+[2023-02-26 09:32:19,532][00001] Num frames 2500...
+[2023-02-26 09:32:19,627][00001] Num frames 2600...
+[2023-02-26 09:32:19,762][00001] Avg episode rewards: #0: 5.480, true rewards: #0: 4.480
+[2023-02-26 09:32:19,762][00001] Avg episode reward: 5.480, avg true_objective: 4.480
+[2023-02-26 09:32:19,773][00001] Num frames 2700...
+[2023-02-26 09:32:19,866][00001] Num frames 2800...
+[2023-02-26 09:32:19,958][00001] Num frames 2900...
+[2023-02-26 09:32:20,051][00001] Avg episode rewards: #0: 5.063, true rewards: #0: 4.206
+[2023-02-26 09:32:20,051][00001] Avg episode reward: 5.063, avg true_objective: 4.206
+[2023-02-26 09:32:20,126][00001] Num frames 3000...
+[2023-02-26 09:32:20,219][00001] Num frames 3100...
+[2023-02-26 09:32:20,312][00001] Num frames 3200...
+[2023-02-26 09:32:20,405][00001] Num frames 3300...
+[2023-02-26 09:32:20,483][00001] Avg episode rewards: #0: 4.910, true rewards: #0: 4.160
+[2023-02-26 09:32:20,483][00001] Avg episode reward: 4.910, avg true_objective: 4.160
+[2023-02-26 09:32:20,575][00001] Num frames 3400...
+[2023-02-26 09:32:20,667][00001] Num frames 3500...
+[2023-02-26 09:32:20,760][00001] Num frames 3600...
+[2023-02-26 09:32:20,853][00001] Num frames 3700...
+[2023-02-26 09:32:20,917][00001] Avg episode rewards: #0: 4.791, true rewards: #0: 4.124
+[2023-02-26 09:32:20,917][00001] Avg episode reward: 4.791, avg true_objective: 4.124
+[2023-02-26 09:32:21,019][00001] Num frames 3800...
+[2023-02-26 09:32:21,112][00001] Num frames 3900...
+[2023-02-26 09:32:21,204][00001] Num frames 4000...
+[2023-02-26 09:32:21,346][00001] Avg episode rewards: #0: 4.696, true rewards: #0: 4.096
+[2023-02-26 09:32:21,346][00001] Avg episode reward: 4.696, avg true_objective: 4.096
+[2023-02-26 09:32:22,553][00001] Replay video saved to /workspace/train_dir/default_experiment/replay.mp4!