Upload folder using huggingface_hub

.summary/0/events.out.tfevents.1694521277.rhmmedcatt-ProLiant-ML350-Gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7de6ffd23dcba7e216952ab00f02a8d679000dec55e57c28732695c82413d334
+size 84366

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_basic
+      type: doom_basic
+    metrics:
+    - type: mean_reward
+      value: 0.77 +/- 0.13
+      name: mean_reward
+      verified: false
+---
+
+A(n) **APPO** model trained on the **doom_basic** 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_basic
+```
+
+    
+## 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_basic --train_dir=./train_dir --experiment=vizdoom_basic
+```
+
+
+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_basic --train_dir=./train_dir --experiment=vizdoom_basic --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_000000224_917504_reward_0.815.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c755a883e08b46f81b73e08a539d7cf4f6181e556acceef80d73a4e936f84f9
+size 34922470

checkpoint_p0/checkpoint_000000246_1007616.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d6187e7885a9977ff2ca58d15928f94edc65c38f207bc232cd061c93bdc675b
+size 34922884

config.json

@@ -0,0 +1,142 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "doom_basic",
+  "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"
+}

sf_log.txt

@@ -0,0 +1,181 @@
+[2023-09-12 13:21:22,562][09743] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:21:22,562][09743] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-09-12 13:21:22,598][09743] Num visible devices: 1
+[2023-09-12 13:21:22,637][09743] Starting seed is not provided
+[2023-09-12 13:21:22,638][09743] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:21:22,638][09743] Initializing actor-critic model on device cuda:0
+[2023-09-12 13:21:22,638][09743] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-12 13:21:22,639][09743] RunningMeanStd input shape: (1,)
+[2023-09-12 13:21:22,659][09743] ConvEncoder: input_channels=3
+[2023-09-12 13:21:22,911][09743] Conv encoder output size: 512
+[2023-09-12 13:21:22,911][09743] Policy head output size: 512
+[2023-09-12 13:21:22,935][09743] Created Actor Critic model with architecture:
+[2023-09-12 13:21:22,935][09743] 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=4, bias=True)
+  )
+)
+[2023-09-12 13:21:24,096][09743] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-09-12 13:21:24,096][09743] No checkpoints found
+[2023-09-12 13:21:24,097][09743] Did not load from checkpoint, starting from scratch!
+[2023-09-12 13:21:24,097][09743] Initialized policy 0 weights for model version 0
+[2023-09-12 13:21:24,098][09743] LearnerWorker_p0 finished initialization!
+[2023-09-12 13:21:24,098][09743] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:21:24,463][09929] Worker 1 uses CPU cores [4, 5, 6, 7]
+[2023-09-12 13:21:24,475][09931] Worker 2 uses CPU cores [8, 9, 10, 11]
+[2023-09-12 13:21:24,499][09964] Worker 5 uses CPU cores [20, 21, 22, 23]
+[2023-09-12 13:21:24,535][09967] Worker 4 uses CPU cores [16, 17, 18, 19]
+[2023-09-12 13:21:24,545][09928] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-09-12 13:21:24,545][09928] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-09-12 13:21:24,566][09928] Num visible devices: 1
+[2023-09-12 13:21:24,567][09932] Worker 3 uses CPU cores [12, 13, 14, 15]
+[2023-09-12 13:21:24,645][09965] Worker 7 uses CPU cores [28, 29, 30, 31]
+[2023-09-12 13:21:24,665][09968] Worker 6 uses CPU cores [24, 25, 26, 27]
+[2023-09-12 13:21:24,689][09930] Worker 0 uses CPU cores [0, 1, 2, 3]
+[2023-09-12 13:21:25,314][09928] RunningMeanStd input shape: (3, 72, 128)
+[2023-09-12 13:21:25,315][09928] RunningMeanStd input shape: (1,)
+[2023-09-12 13:21:25,326][09928] ConvEncoder: input_channels=3
+[2023-09-12 13:21:25,447][09928] Conv encoder output size: 512
+[2023-09-12 13:21:25,448][09928] Policy head output size: 512
+[2023-09-12 13:21:25,839][09964] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:25,839][09968] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:25,839][09967] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:25,840][09965] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:25,840][09931] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:25,848][09932] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:25,851][09930] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:25,852][09929] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-09-12 13:21:26,147][09967] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,147][09965] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,215][09964] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,239][09929] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,258][09968] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,273][09931] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,286][09930] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,418][09967] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:26,493][09964] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:26,522][09965] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:26,525][09929] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:26,551][09932] Decorrelating experience for 0 frames...
+[2023-09-12 13:21:26,556][09931] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:26,568][09930] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:26,775][09967] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:26,821][09932] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:26,852][09964] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:26,919][09931] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:26,929][09930] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:27,103][09929] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:27,160][09968] Decorrelating experience for 32 frames...
+[2023-09-12 13:21:27,164][09965] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:27,195][09932] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:27,201][09964] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:27,330][09931] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:27,451][09929] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:27,465][09967] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:27,498][09965] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:27,507][09930] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:27,588][09968] Decorrelating experience for 64 frames...
+[2023-09-12 13:21:27,634][09932] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:27,903][09968] Decorrelating experience for 96 frames...
+[2023-09-12 13:21:28,649][09743] Signal inference workers to stop experience collection...
+[2023-09-12 13:21:28,653][09928] InferenceWorker_p0-w0: stopping experience collection
+[2023-09-12 13:21:32,650][09743] Signal inference workers to resume experience collection...
+[2023-09-12 13:21:32,651][09928] InferenceWorker_p0-w0: resuming experience collection
+[2023-09-12 13:21:35,991][09928] Updated weights for policy 0, policy_version 10 (0.0392)
+[2023-09-12 13:21:39,213][09928] Updated weights for policy 0, policy_version 20 (0.0009)
+[2023-09-12 13:21:42,363][09928] Updated weights for policy 0, policy_version 30 (0.0009)
+[2023-09-12 13:21:42,527][09743] Saving new best policy, reward=-1.655!
+[2023-09-12 13:21:45,525][09928] Updated weights for policy 0, policy_version 40 (0.0009)
+[2023-09-12 13:21:47,532][09743] Saving new best policy, reward=-0.936!
+[2023-09-12 13:21:48,749][09928] Updated weights for policy 0, policy_version 50 (0.0009)
+[2023-09-12 13:21:51,927][09928] Updated weights for policy 0, policy_version 60 (0.0008)
+[2023-09-12 13:21:52,564][09743] Saving new best policy, reward=0.078!
+[2023-09-12 13:21:55,197][09928] Updated weights for policy 0, policy_version 70 (0.0009)
+[2023-09-12 13:21:57,529][09743] Saving new best policy, reward=0.521!
+[2023-09-12 13:21:58,483][09928] Updated weights for policy 0, policy_version 80 (0.0010)
+[2023-09-12 13:22:01,740][09928] Updated weights for policy 0, policy_version 90 (0.0008)
+[2023-09-12 13:22:02,527][09743] Saving new best policy, reward=0.599!
+[2023-09-12 13:22:05,213][09928] Updated weights for policy 0, policy_version 100 (0.0012)
+[2023-09-12 13:22:07,589][09743] Saving new best policy, reward=0.680!
+[2023-09-12 13:22:08,636][09928] Updated weights for policy 0, policy_version 110 (0.0017)
+[2023-09-12 13:22:11,996][09928] Updated weights for policy 0, policy_version 120 (0.0009)
+[2023-09-12 13:22:12,526][09743] Saving new best policy, reward=0.735!
+[2023-09-12 13:22:15,370][09928] Updated weights for policy 0, policy_version 130 (0.0008)
+[2023-09-12 13:22:17,527][09743] Saving new best policy, reward=0.755!
+[2023-09-12 13:22:18,771][09928] Updated weights for policy 0, policy_version 140 (0.0009)
+[2023-09-12 13:22:22,142][09928] Updated weights for policy 0, policy_version 150 (0.0009)
+[2023-09-12 13:22:22,526][09743] Saving new best policy, reward=0.781!
+[2023-09-12 13:22:25,580][09928] Updated weights for policy 0, policy_version 160 (0.0008)
+[2023-09-12 13:22:27,573][09743] Saving new best policy, reward=0.791!
+[2023-09-12 13:22:28,937][09928] Updated weights for policy 0, policy_version 170 (0.0009)
+[2023-09-12 13:22:32,402][09928] Updated weights for policy 0, policy_version 180 (0.0008)
+[2023-09-12 13:22:32,526][09743] Saving new best policy, reward=0.794!
+[2023-09-12 13:22:35,725][09928] Updated weights for policy 0, policy_version 190 (0.0009)
+[2023-09-12 13:22:37,529][09743] Saving new best policy, reward=0.806!
+[2023-09-12 13:22:39,128][09928] Updated weights for policy 0, policy_version 200 (0.0010)
+[2023-09-12 13:22:42,471][09928] Updated weights for policy 0, policy_version 210 (0.0009)
+[2023-09-12 13:22:45,922][09928] Updated weights for policy 0, policy_version 220 (0.0009)
+[2023-09-12 13:22:47,533][09743] Saving new best policy, reward=0.815!
+[2023-09-12 13:22:49,374][09928] Updated weights for policy 0, policy_version 230 (0.0009)
+[2023-09-12 13:22:52,742][09928] Updated weights for policy 0, policy_version 240 (0.0009)
+[2023-09-12 13:22:54,862][09743] Stopping Batcher_0...
+[2023-09-12 13:22:54,862][09743] Saving /home/cogstack/Documents/optuna/environments/sample_factory/train_dir/default_experiment/checkpoint_p0/checkpoint_000000246_1007616.pth...
+[2023-09-12 13:22:54,863][09743] Loop batcher_evt_loop terminating...
+[2023-09-12 13:22:54,876][09965] Stopping RolloutWorker_w7...
+[2023-09-12 13:22:54,877][09965] Loop rollout_proc7_evt_loop terminating...
+[2023-09-12 13:22:54,877][09932] Stopping RolloutWorker_w3...
+[2023-09-12 13:22:54,877][09930] Stopping RolloutWorker_w0...
+[2023-09-12 13:22:54,877][09968] Stopping RolloutWorker_w6...
+[2023-09-12 13:22:54,877][09932] Loop rollout_proc3_evt_loop terminating...
+[2023-09-12 13:22:54,877][09930] Loop rollout_proc0_evt_loop terminating...
+[2023-09-12 13:22:54,878][09968] Loop rollout_proc6_evt_loop terminating...
+[2023-09-12 13:22:54,880][09964] Stopping RolloutWorker_w5...
+[2023-09-12 13:22:54,880][09931] Stopping RolloutWorker_w2...
+[2023-09-12 13:22:54,880][09964] Loop rollout_proc5_evt_loop terminating...
+[2023-09-12 13:22:54,880][09931] Loop rollout_proc2_evt_loop terminating...
+[2023-09-12 13:22:54,881][09929] Stopping RolloutWorker_w1...
+[2023-09-12 13:22:54,881][09929] Loop rollout_proc1_evt_loop terminating...
+[2023-09-12 13:22:54,882][09967] Stopping RolloutWorker_w4...
+[2023-09-12 13:22:54,882][09967] Loop rollout_proc4_evt_loop terminating...
+[2023-09-12 13:22:54,885][09928] Weights refcount: 2 0
+[2023-09-12 13:22:54,887][09928] Stopping InferenceWorker_p0-w0...
+[2023-09-12 13:22:54,887][09928] Loop inference_proc0-0_evt_loop terminating...
+[2023-09-12 13:22:54,931][09743] Saving /home/cogstack/Documents/optuna/environments/sample_factory/train_dir/default_experiment/checkpoint_p0/checkpoint_000000246_1007616.pth...
+[2023-09-12 13:22:55,021][09743] Stopping LearnerWorker_p0...
+[2023-09-12 13:22:55,022][09743] Loop learner_proc0_evt_loop terminating...