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.1694522070.rhmmedcatt-ProLiant-ML350-Gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e805eb9d3bbfa748e20217eef5ebff324849ec424fe293164cf3f77ef3703353
+size 89802

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_two_colors_easy
+      type: doom_two_colors_easy
+    metrics:
+    - type: mean_reward
+      value: 3.90 +/- 0.59
+      name: mean_reward
+      verified: false
+---
+
+A(n) **APPO** model trained on the **doom_two_colors_easy** 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 MattStammers/vizdoom_two_colors_easy
+```
+
+    
+## 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_two_colors_easy --train_dir=./train_dir --experiment=vizdoom_two_colors_easy
+```
+
+
+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_two_colors_easy --train_dir=./train_dir --experiment=vizdoom_two_colors_easy --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_000000027_110592_reward_4.631.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5885fae74895f6651020bc70c248ffc3cffae91d90f7c1e31b5f3229653afb71
+size 34934758

checkpoint_p0/checkpoint_000000246_1007616.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d540c7e0ce330e4dd41d326104cdcb9328631893b95ac2c1eb524f89b997a642
+size 34935172

config.json

@@ -0,0 +1,142 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "doom_two_colors_easy",
+  "experiment": "default_experiment",
+  "train_dir": "/home/cogstack/Documents/optuna/environments/sample_factory/train_dir",
+  "restart_behavior": "restart",
+  "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": 1000000,
+  "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=4000000",
+  "cli_args": {
+    "env": "doom_health_gathering_supreme",
+    "num_workers": 8,
+    "num_envs_per_worker": 4,
+    "train_for_env_steps": 4000000
+  },
+  "git_hash": "b12d96985caa7a7552d0840afdd14065f56f9f9a",
+  "git_repo_name": "https://github.com/MattStammers/optuna.git"
+}

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe1a058d390ea968a6df89a22a015e13f107c5b56e2f562f4f7f2cc02b0a5fcc
+size 4888196

sf_log.txt

@@ -0,0 +1,169 @@
+[2023-09-12 13:34:35,213][63361] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:34:35,214][63361] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-09-12 13:34:35,232][63361] Num visible devices: 1
+[2023-09-12 13:34:35,270][63361] Starting seed is not provided
+[2023-09-12 13:34:35,270][63361] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:34:35,270][63361] Initializing actor-critic model on device cuda:0
+[2023-09-12 13:34:35,271][63361] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-12 13:34:35,271][63361] RunningMeanStd input shape: (1,)
+[2023-09-12 13:34:35,284][63361] ConvEncoder: input_channels=3
+[2023-09-12 13:34:35,443][63361] Conv encoder output size: 512
+[2023-09-12 13:34:35,443][63361] Policy head output size: 512
+[2023-09-12 13:34:35,467][63361] Created Actor Critic model with architecture:
+[2023-09-12 13:34:35,467][63361] 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=6, bias=True)
+  )
+)
+[2023-09-12 13:34:36,777][63361] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-09-12 13:34:36,778][63361] No checkpoints found
+[2023-09-12 13:34:36,778][63361] Did not load from checkpoint, starting from scratch!
+[2023-09-12 13:34:36,778][63361] Initialized policy 0 weights for model version 0
+[2023-09-12 13:34:36,779][63361] LearnerWorker_p0 finished initialization!
+[2023-09-12 13:34:36,780][63361] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:34:37,202][63596] Worker 7 uses CPU cores [28, 29, 30, 31]
+[2023-09-12 13:34:37,203][63553] Worker 1 uses CPU cores [4, 5, 6, 7]
+[2023-09-12 13:34:37,244][63587] Worker 0 uses CPU cores [0, 1, 2, 3]
+[2023-09-12 13:34:37,255][63595] Worker 6 uses CPU cores [24, 25, 26, 27]
+[2023-09-12 13:34:37,259][63594] Worker 5 uses CPU cores [20, 21, 22, 23]
+[2023-09-12 13:34:37,359][63586] Worker 3 uses CPU cores [12, 13, 14, 15]
+[2023-09-12 13:34:37,367][63590] Worker 2 uses CPU cores [8, 9, 10, 11]
+[2023-09-12 13:34:37,367][63593] Worker 4 uses CPU cores [16, 17, 18, 19]
+[2023-09-12 13:34:37,406][63552] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:34:37,406][63552] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-09-12 13:34:37,425][63552] Num visible devices: 1
+[2023-09-12 13:34:38,097][63552] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-12 13:34:38,098][63552] RunningMeanStd input shape: (1,)
+[2023-09-12 13:34:38,109][63552] ConvEncoder: input_channels=3
+[2023-09-12 13:34:38,210][63552] Conv encoder output size: 512
+[2023-09-12 13:34:38,210][63552] Policy head output size: 512
+[2023-09-12 13:34:38,579][63596] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,585][63595] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,585][63593] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,585][63587] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,595][63586] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,597][63590] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,604][63553] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,604][63594] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:34:38,878][63596] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:38,895][63593] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:38,900][63595] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:38,922][63586] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:39,153][63587] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:39,181][63593] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,182][63595] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,228][63596] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,229][63594] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:39,231][63586] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,254][63590] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:39,500][63594] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,513][63587] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,523][63590] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,525][63553] Decorrelating experience for 0 frames...
+[2023-09-12 13:34:39,533][63593] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:39,788][63553] Decorrelating experience for 32 frames...
+[2023-09-12 13:34:39,838][63595] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:39,843][63586] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:39,849][63593] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:39,863][63587] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:40,155][63595] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:40,176][63587] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:40,218][63553] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:40,221][63594] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:40,247][63586] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:40,499][63590] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:40,609][63594] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:40,618][63553] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:40,838][63596] Decorrelating experience for 64 frames...
+[2023-09-12 13:34:41,027][63590] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:41,229][63596] Decorrelating experience for 96 frames...
+[2023-09-12 13:34:41,428][63361] Signal inference workers to stop experience collection...
+[2023-09-12 13:34:41,432][63552] InferenceWorker_p0-w0: stopping experience collection
+[2023-09-12 13:34:45,435][63361] Signal inference workers to resume experience collection...
+[2023-09-12 13:34:45,436][63552] InferenceWorker_p0-w0: resuming experience collection
+[2023-09-12 13:34:48,988][63552] Updated weights for policy 0, policy_version 10 (0.0376)
+[2023-09-12 13:34:52,732][63552] Updated weights for policy 0, policy_version 20 (0.0010)
+[2023-09-12 13:34:55,247][63361] Saving new best policy, reward=4.631!
+[2023-09-12 13:34:56,268][63552] Updated weights for policy 0, policy_version 30 (0.0009)
+[2023-09-12 13:34:59,869][63552] Updated weights for policy 0, policy_version 40 (0.0009)
+[2023-09-12 13:35:03,454][63552] Updated weights for policy 0, policy_version 50 (0.0009)
+[2023-09-12 13:35:07,103][63552] Updated weights for policy 0, policy_version 60 (0.0009)
+[2023-09-12 13:35:10,657][63552] Updated weights for policy 0, policy_version 70 (0.0009)
+[2023-09-12 13:35:14,191][63552] Updated weights for policy 0, policy_version 80 (0.0009)
+[2023-09-12 13:35:17,722][63552] Updated weights for policy 0, policy_version 90 (0.0009)
+[2023-09-12 13:35:21,228][63552] Updated weights for policy 0, policy_version 100 (0.0008)
+[2023-09-12 13:35:24,812][63552] Updated weights for policy 0, policy_version 110 (0.0009)
+[2023-09-12 13:35:28,361][63552] Updated weights for policy 0, policy_version 120 (0.0008)
+[2023-09-12 13:35:31,821][63552] Updated weights for policy 0, policy_version 130 (0.0009)
+[2023-09-12 13:35:35,276][63552] Updated weights for policy 0, policy_version 140 (0.0008)
+[2023-09-12 13:35:38,747][63552] Updated weights for policy 0, policy_version 150 (0.0008)
+[2023-09-12 13:35:42,301][63552] Updated weights for policy 0, policy_version 160 (0.0008)
+[2023-09-12 13:35:45,805][63552] Updated weights for policy 0, policy_version 170 (0.0008)
+[2023-09-12 13:35:49,340][63552] Updated weights for policy 0, policy_version 180 (0.0008)
+[2023-09-12 13:35:52,925][63552] Updated weights for policy 0, policy_version 190 (0.0008)
+[2023-09-12 13:35:56,545][63552] Updated weights for policy 0, policy_version 200 (0.0008)
+[2023-09-12 13:36:00,090][63552] Updated weights for policy 0, policy_version 210 (0.0009)
+[2023-09-12 13:36:03,599][63552] Updated weights for policy 0, policy_version 220 (0.0009)
+[2023-09-12 13:36:07,145][63552] Updated weights for policy 0, policy_version 230 (0.0008)
+[2023-09-12 13:36:10,690][63552] Updated weights for policy 0, policy_version 240 (0.0008)
+[2023-09-12 13:36:12,775][63361] Saving /home/cogstack/Documents/optuna/environments/sample_factory/train_dir/default_experiment/checkpoint_p0/checkpoint_000000246_1007616.pth...
+[2023-09-12 13:36:12,789][63593] Stopping RolloutWorker_w4...
+[2023-09-12 13:36:12,789][63593] Loop rollout_proc4_evt_loop terminating...
+[2023-09-12 13:36:12,790][63590] Stopping RolloutWorker_w2...
+[2023-09-12 13:36:12,790][63590] Loop rollout_proc2_evt_loop terminating...
+[2023-09-12 13:36:12,791][63595] Stopping RolloutWorker_w6...
+[2023-09-12 13:36:12,791][63595] Loop rollout_proc6_evt_loop terminating...
+[2023-09-12 13:36:12,792][63587] Stopping RolloutWorker_w0...
+[2023-09-12 13:36:12,792][63596] Stopping RolloutWorker_w7...
+[2023-09-12 13:36:12,793][63587] Loop rollout_proc0_evt_loop terminating...
+[2023-09-12 13:36:12,793][63596] Loop rollout_proc7_evt_loop terminating...
+[2023-09-12 13:36:12,794][63586] Stopping RolloutWorker_w3...
+[2023-09-12 13:36:12,794][63586] Loop rollout_proc3_evt_loop terminating...
+[2023-09-12 13:36:12,786][63361] Stopping Batcher_0...
+[2023-09-12 13:36:12,795][63594] Stopping RolloutWorker_w5...
+[2023-09-12 13:36:12,795][63594] Loop rollout_proc5_evt_loop terminating...
+[2023-09-12 13:36:12,796][63552] Weights refcount: 2 0
+[2023-09-12 13:36:12,797][63552] Stopping InferenceWorker_p0-w0...
+[2023-09-12 13:36:12,798][63552] Loop inference_proc0-0_evt_loop terminating...
+[2023-09-12 13:36:12,807][63553] Stopping RolloutWorker_w1...
+[2023-09-12 13:36:12,807][63553] Loop rollout_proc1_evt_loop terminating...
+[2023-09-12 13:36:12,803][63361] Loop batcher_evt_loop terminating...
+[2023-09-12 13:36:12,828][63361] Saving /home/cogstack/Documents/optuna/environments/sample_factory/train_dir/default_experiment/checkpoint_p0/checkpoint_000000246_1007616.pth...
+[2023-09-12 13:36:12,891][63361] Stopping LearnerWorker_p0...
+[2023-09-12 13:36:12,891][63361] Loop learner_proc0_evt_loop terminating...