Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,6 @@ 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
+env_a100_100x100.mp4 filter=lfs diff=lfs merge=lfs -text
+env_a20_40x40.mp4 filter=lfs diff=lfs merge=lfs -text
+env_a5_25x25.mp4 filter=lfs diff=lfs merge=lfs -text

.summary/0/events.out.tfevents.1711738851.ip-172-31-37-173.ec2.internal

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17b2c436da32752187fe6891765864e61624730cc1abb8329640186b37744380
+size 3837593

.summary/0/events.out.tfevents.1711742223.ip-172-31-79-185.ec2.internal

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:622cb2f3fe692d1e681398f4a21706e111b93e42a293243c0609c1d7b2aba9b5
+size 12132412

README.md

@@ -0,0 +1,42 @@
+---
+library_name: sample-factory
+tags:
+- deep-reinforcement-learning
+- reinforcement-learning
+- sample-factory
+---
+
+A(n) **APPO** model trained on the **GDY-PowerGrid** 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 metta-ai/baseline.v0.0.3
+```
+
+    
+## 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=GDY-PowerGrid --train_dir=./train_dir --experiment=baseline.v0.0.3
+```
+
+
+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=GDY-PowerGrid --train_dir=./train_dir --experiment=baseline.v0.0.3 --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_000034523_565624832_reward_0.655.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:130967a109d0b90992c36785f87c8ae209f69f089993a25fd6f31034ccd7ea05
+size 76478202

checkpoint_p0/checkpoint_000040205_658718720.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f94b25afafd91be3f52a2e404a99f28e2c0a18683961f8f13f12f1c15272adaa
+size 76479020

checkpoint_p0/checkpoint_000040513_663764992.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14c3a1ed44dbcd7cb5506562f51f5cb5a36fe62683f79c26c410932898fe1b30
+size 76479020

config.json

@@ -0,0 +1,322 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "GDY-PowerGrid",
+  "experiment": "b.a20.20x20_40x40.norm",
+  "train_dir": "/workspace/metta/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": 50,
+  "num_workers": 64,
+  "num_envs_per_worker": 2,
+  "batch_size": 16384,
+  "num_batches_per_epoch": 1,
+  "num_epochs": 1,
+  "rollout": 256,
+  "recurrence": 256,
+  "shuffle_minibatches": false,
+  "gamma": 0.99,
+  "reward_scale": 1.0,
+  "reward_clip": 1000.0,
+  "value_bootstrap": false,
+  "normalize_returns": true,
+  "exploration_loss_coeff": 0.002,
+  "value_loss_coeff": 0.976,
+  "kl_loss_coeff": 0.0,
+  "aux_loss_coeff": 0.0,
+  "exploration_loss": "symmetric_kl",
+  "gae_lambda": 0.95,
+  "ppo_clip_ratio": 0.1,
+  "ppo_clip_value": 1.0,
+  "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": 1.0,
+  "normalize_input": true,
+  "normalize_input_keys": null,
+  "decorrelate_experience_max_seconds": 150,
+  "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": 180,
+  "train_for_env_steps": 10000000000,
+  "train_for_seconds": 10000000000,
+  "save_every_sec": 120,
+  "keep_checkpoints": 2,
+  "load_checkpoint_path": null,
+  "init_checkpoint_path": null,
+  "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": 1,
+  "env_framestack": 1,
+  "pixel_format": "CHW",
+  "use_record_episode_statistics": false,
+  "episode_counter": false,
+  "with_wandb": true,
+  "wandb_user": "platypus",
+  "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,
+  "env_altar:cooldown": [
+    2,
+    5
+  ],
+  "env_altar:cost": [
+    100.0
+  ],
+  "env_charger:cooldown": [
+    2,
+    2
+  ],
+  "env_charger:energy": [
+    50,
+    50
+  ],
+  "env_generator:cooldown": [
+    20,
+    100
+  ],
+  "env_agent:energy:regen": [
+    0,
+    0
+  ],
+  "env_agent:energy:initial": [
+    10,
+    200
+  ],
+  "env_agent:energy:max": [
+    500,
+    500
+  ],
+  "env_gift:energy": [
+    20,
+    20
+  ],
+  "env_cost:move:predator": [
+    0,
+    0
+  ],
+  "env_cost:move:prey": [
+    0,
+    0
+  ],
+  "env_cost:move": [
+    0,
+    0
+  ],
+  "env_cost:jump": [
+    3,
+    3
+  ],
+  "env_cost:rotate": [
+    0,
+    0
+  ],
+  "env_cost:shield": [
+    0.0
+  ],
+  "env_cost:shield:upkeep": [
+    1,
+    2
+  ],
+  "env_cost:frozen": [
+    0,
+    0
+  ],
+  "env_cost:attack": [
+    5,
+    40
+  ],
+  "env_attack:damage": [
+    5,
+    40
+  ],
+  "env_attack:freeze_duration": [
+    5,
+    100
+  ],
+  "env_num_agents": [
+    20.0
+  ],
+  "env_tile_size": [
+    32,
+    32
+  ],
+  "env_width": [
+    20.0,
+    40.0
+  ],
+  "env_height": [
+    20.0,
+    40.0
+  ],
+  "env_max_steps": [
+    1000,
+    1000
+  ],
+  "env_wall_density": [
+    0.01
+  ],
+  "env_milestone_density": [
+    0,
+    0.1
+  ],
+  "env_num_altars": [
+    5.0
+  ],
+  "env_num_chargers": [
+    10.0
+  ],
+  "env_num_generators": [
+    50.0
+  ],
+  "env_rsm_num_families": [
+    0,
+    0
+  ],
+  "env_rsm_family_reward": [
+    0,
+    0
+  ],
+  "env_reward_rank_steps": [
+    1000,
+    1000
+  ],
+  "env_reward_prestige_weight": [
+    0,
+    0
+  ],
+  "agent_max_features": 50,
+  "agent_fc_layers": 3,
+  "agent_fc_size": 512,
+  "agent_embedding_size": 64,
+  "agent_embedding_layers": 3,
+  "command_line": "--env_wall_density=0.01 --env_altar:cost=100 --env_cost:shield=0 --env_num_agents=20 --env_width=20:40 --env_height=20:40 --env_num_altars=5 --env_num_chargers=10 --env_num_generators=50 --normalize_input=True --aux_loss_coef=0 --recurrence=256 --rollout=256 --value_loss_coeff=0.976 --exploration_loss=symmetric_kl --exploration_loss_coeff=0.002 --policy_initialization=orthogonal --learning_rate=0.0001 --max_policy_lag=50 --nonlinearity=elu --load_checkpoint_kind=latest --rnn_num_layers=1 --rnn_size=512 --rnn_type=gru --agent_fc_layers=3 --agent_fc_size=512 --agent_embedding_size=64 --agent_embedding_layers=3 --with_wandb=True --wandb_user=platypus --wandb_project=sample_factory --batch_size=16384 --decorrelate_experience_max_seconds=150 --experiment=b.a20.20x20_40x40.norm --env=GDY-PowerGrid",
+  "cli_args": {
+    "env": "GDY-PowerGrid",
+    "experiment": "b.a20.20x20_40x40.norm",
+    "max_policy_lag": 50,
+    "batch_size": 16384,
+    "rollout": 256,
+    "recurrence": 256,
+    "exploration_loss_coeff": 0.002,
+    "value_loss_coeff": 0.976,
+    "aux_loss_coeff": 0.0,
+    "exploration_loss": "symmetric_kl",
+    "learning_rate": 0.0001,
+    "normalize_input": true,
+    "decorrelate_experience_max_seconds": 150,
+    "load_checkpoint_kind": "latest",
+    "rnn_size": 512,
+    "rnn_type": "gru",
+    "rnn_num_layers": 1,
+    "nonlinearity": "elu",
+    "policy_initialization": "orthogonal",
+    "with_wandb": true,
+    "wandb_user": "platypus",
+    "wandb_project": "sample_factory",
+    "env_altar:cost": [
+      100.0
+    ],
+    "env_cost:shield": [
+      0.0
+    ],
+    "env_num_agents": [
+      20.0
+    ],
+    "env_width": [
+      20.0,
+      40.0
+    ],
+    "env_height": [
+      20.0,
+      40.0
+    ],
+    "env_wall_density": [
+      0.01
+    ],
+    "env_num_altars": [
+      5.0
+    ],
+    "env_num_chargers": [
+      10.0
+    ],
+    "env_num_generators": [
+      50.0
+    ],
+    "agent_fc_layers": 3,
+    "agent_fc_size": 512,
+    "agent_embedding_size": 64,
+    "agent_embedding_layers": 3
+  },
+  "git_hash": "a9d0c93c1854a34c7dd674bf327bc59b9831e10a",
+  "git_repo_name": "https://github.com/daveey/metta.git",
+  "wandb_unique_id": "b.a20.20x20_40x40.norm_20240329_120046_450296"
+}

env_a100_100x100.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7467609f18ba03fdf28e28d4b77fcce5c6bb15cd0ef13a7f2436c9fc646cde5d
+size 12425790

env_a20_40x40.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d6ac88750b89110511d737c7040357fff3569f5100ec39dc3ea9e74bfadc211
+size 6493147

env_a5_25x25.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97b126c88688ac6cd1fecf73e5951ddcca4793f12e02c801c55e407a58e65c6d
+size 2584700

git.diff

@@ -0,0 +1,7 @@
+diff --git a/third_party/sample_factory b/third_party/sample_factory
+index dfcb7ba..f7d3662 160000
+--- a/third_party/sample_factory
++++ b/third_party/sample_factory
+@@ -1 +1 @@
+-Subproject commit dfcb7ba7274f4867f6d3238d7ea0ec84ddb5c28f
++Subproject commit f7d36626407b6e061fec0677a1e85931c17cdaef