Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,7 @@ 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
+a20_40x40.mp4 filter=lfs diff=lfs merge=lfs -text
+a20_b4_40x40.mp4 filter=lfs diff=lfs merge=lfs -text
+a20_r4_40x40.mp4 filter=lfs diff=lfs merge=lfs -text
+replay.mp4 filter=lfs diff=lfs merge=lfs -text

.summary/0/events.out.tfevents.1718036794.c11948e043d9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c303278fbbc65af492365425320a83593d8245396b57467da29f4025e45f98f
+size 8132944

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-MettaGrid** 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.v0.3.1
+```
+
+    
+## 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-MettaGrid --train_dir=./train_dir --experiment=baseline.v0.3.1
+```
+
+
+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-MettaGrid --train_dir=./train_dir --experiment=baseline.v0.3.1 --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.
+    

a20_40x40.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43bb854413b30feb1e4a61d81b0ab8ac25a0b9252d3a3d2129bf617f4dfc84f1
+size 1963979

a20_b4_40x40.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd636514e9de0f199d5517ebab91434743aca4e2954360a3a36ee313e54bc317
+size 1603862

a20_r4_40x40.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f2cb9a7d21f3952be7aa8e8f629b18989773a42ad87535efcbf5b0cc19b46bd
+size 2214751

checkpoint_p0/best_000022096_362020864_reward_0.299.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f54e627ebfc1b09f4e4bb237a0d47983adb2c58e523662453321e780185e92cb
+size 38972794

checkpoint_p0/checkpoint_000023879_391233536.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a5f4ce4badd628eca7450fb87af3b04909680ffa6ca127f290875a8f6b6a819
+size 38974102

checkpoint_p0/checkpoint_000024206_396591104.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:031303a811c9a1ad49ec1b88df5503e996932a1e623bf7187fa82b118cdab4ee
+size 38974102

config.json

@@ -0,0 +1,161 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "GDY-MettaGrid",
+  "experiment": "p2.metta.6",
+  "train_dir": "/workspace/metta/train_dir",
+  "restart_behavior": "resume",
+  "device": "gpu",
+  "seed": 0,
+  "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": 32,
+  "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": false,
+  "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": 9999999999999,
+  "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": "metta",
+  "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_cfg": "{\"name\": \"GDY-MettaGrid\", \"_target_\": \"env.griddly.mettagrid.gym_env.MettaGridGymEnv\", \"max_action_value\": 10, \"game_builder\": {\"_target_\": \"env.griddly.mettagrid.game_builder.MettaGridGameBuilder\", \"obs_width\": 11, \"obs_height\": 11, \"max_steps\": 1000, \"tile_size\": 16, \"num_agents\": 20, \"objects\": {\"agent\": {\"initial_energy\": 100, \"max_energy\": 500, \"max_inventory\": 5, \"freeze_duration\": 10, \"hp\": 1, \"upkeep\": {\"shield\": 1}}, \"altar\": {\"hp\": 30, \"cooldown\": 2, \"cost\": 100}, \"converter\": {\"hp\": 30, \"cooldown\": 2, \"energy_output\": 50, \"cost\": 0}, \"generator\": {\"hp\": 30, \"cooldown\": 5, \"initial_resources\": 30, \"cost\": 0}, \"wall\": {\"density\": 0.01, \"hp\": 10}}, \"actions\": {\"move\": {\"cost\": 0}, \"rotate\": {\"cost\": 0}, \"jump\": {\"cost\": 3}, \"shield\": {\"cost\": 0}, \"drop\": {\"cost\": 1}, \"use\": {\"cost\": 0}, \"attack\": {\"cost\": 5, \"damage\": 5}}, \"map\": {\"layout\": [[\"base\", \"wild_1\", \"base\"], [\"wild_2\", \"center\", \"wild_2\"], [\"base\", \"wild_1\", \"base\"]], \"room\": {\"width\": 25, \"height\": 25, \"num_agents\": 5, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 15, \"wall\": 10}}, \"wild_1\": {\"width\": 10, \"height\": 15, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 1, \"converter\": 1, \"generator\": 5, \"wall\": 5}}, \"wild_2\": {\"width\": 15, \"height\": 10, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 1, \"converter\": 1, \"generator\": 5, \"wall\": 5}}, \"center\": {\"width\": 10, \"height\": 10, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 2, \"converter\": 5, \"generator\": 10, \"wall\": 20}}, \"base\": {\"width\": 15, \"height\": 15, \"border\": 1, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 5, \"wall\": 5}}}}}",
+  "agent_cfg": "{\"_target_\": \"agent.metta_agent.MettaAgent\", \"observation_encoders\": {\"grid_obs\": {\"feature_names\": [], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 512, \"layers\": 4}, \"global_vars\": {\"feature_names\": [], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}, \"last_action\": {\"feature_names\": [\"last_action_id\", \"last_action_val\"], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}, \"last_reward\": {\"feature_names\": [\"last_reward\"], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}}, \"fc\": {\"layers\": 3, \"output_dim\": 512}, \"decoder\": {\"_target_\": \"agent.decoder.Decoder\"}, \"core\": {\"_target_\": \"sample_factory.model.core.ModelCoreRNN\", \"rnn_type\": \"gru\", \"rnn_num_layers\": 1, \"rnn_size\": 512}}",
+  "command_line": "--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 --normalize_input=False --seed=0 --batch_size=16384 --decorrelate_experience_max_seconds=150 --train_for_env_steps=9999999999999 --with_wandb=True --wandb_user=platypus --wandb_project=metta --experiment=p2.metta.6 --rnn_type=gru --rnn_num_layers=1 --rnn_size=512 --env=GDY-MettaGrid --env_cfg={\"name\": \"GDY-MettaGrid\", \"_target_\": \"env.griddly.mettagrid.gym_env.MettaGridGymEnv\", \"max_action_value\": 10, \"game_builder\": {\"_target_\": \"env.griddly.mettagrid.game_builder.MettaGridGameBuilder\", \"obs_width\": 11, \"obs_height\": 11, \"max_steps\": 1000, \"tile_size\": 16, \"num_agents\": 20, \"objects\": {\"agent\": {\"initial_energy\": 100, \"max_energy\": 500, \"max_inventory\": 5, \"freeze_duration\": 10, \"hp\": 1, \"upkeep\": {\"shield\": 1}}, \"altar\": {\"hp\": 30, \"cooldown\": 2, \"cost\": 100}, \"converter\": {\"hp\": 30, \"cooldown\": 2, \"energy_output\": 50, \"cost\": 0}, \"generator\": {\"hp\": 30, \"cooldown\": 5, \"initial_resources\": 30, \"cost\": 0}, \"wall\": {\"density\": 0.01, \"hp\": 10}}, \"actions\": {\"move\": {\"cost\": 0}, \"rotate\": {\"cost\": 0}, \"jump\": {\"cost\": 3}, \"shield\": {\"cost\": 0}, \"drop\": {\"cost\": 1}, \"use\": {\"cost\": 0}, \"attack\": {\"cost\": 5, \"damage\": 5}}, \"map\": {\"layout\": [[\"base\", \"wild_1\", \"base\"], [\"wild_2\", \"center\", \"wild_2\"], [\"base\", \"wild_1\", \"base\"]], \"room\": {\"width\": 25, \"height\": 25, \"num_agents\": 5, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 15, \"wall\": 10}}, \"wild_1\": {\"width\": 10, \"height\": 15, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 1, \"converter\": 1, \"generator\": 5, \"wall\": 5}}, \"wild_2\": {\"width\": 15, \"height\": 10, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 1, \"converter\": 1, \"generator\": 5, \"wall\": 5}}, \"center\": {\"width\": 10, \"height\": 10, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 2, \"converter\": 5, \"generator\": 10, \"wall\": 20}}, \"base\": {\"width\": 15, \"height\": 15, \"border\": 1, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 5, \"wall\": 5}}}}} --agent_cfg={\"_target_\": \"agent.metta_agent.MettaAgent\", \"observation_encoders\": {\"grid_obs\": {\"feature_names\": [], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 512, \"layers\": 4}, \"global_vars\": {\"feature_names\": [], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}, \"last_action\": {\"feature_names\": [\"last_action_id\", \"last_action_val\"], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}, \"last_reward\": {\"feature_names\": [\"last_reward\"], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}}, \"fc\": {\"layers\": 3, \"output_dim\": 512}, \"decoder\": {\"_target_\": \"agent.decoder.Decoder\"}, \"core\": {\"_target_\": \"sample_factory.model.core.ModelCoreRNN\", \"rnn_type\": \"gru\", \"rnn_num_layers\": 1, \"rnn_size\": 512}}",
+  "cli_args": {
+    "env": "GDY-MettaGrid",
+    "experiment": "p2.metta.6",
+    "seed": 0,
+    "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": false,
+    "decorrelate_experience_max_seconds": 150,
+    "train_for_env_steps": 9999999999999,
+    "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": "metta",
+    "env_cfg": "{\"name\": \"GDY-MettaGrid\", \"_target_\": \"env.griddly.mettagrid.gym_env.MettaGridGymEnv\", \"max_action_value\": 10, \"game_builder\": {\"_target_\": \"env.griddly.mettagrid.game_builder.MettaGridGameBuilder\", \"obs_width\": 11, \"obs_height\": 11, \"max_steps\": 1000, \"tile_size\": 16, \"num_agents\": 20, \"objects\": {\"agent\": {\"initial_energy\": 100, \"max_energy\": 500, \"max_inventory\": 5, \"freeze_duration\": 10, \"hp\": 1, \"upkeep\": {\"shield\": 1}}, \"altar\": {\"hp\": 30, \"cooldown\": 2, \"cost\": 100}, \"converter\": {\"hp\": 30, \"cooldown\": 2, \"energy_output\": 50, \"cost\": 0}, \"generator\": {\"hp\": 30, \"cooldown\": 5, \"initial_resources\": 30, \"cost\": 0}, \"wall\": {\"density\": 0.01, \"hp\": 10}}, \"actions\": {\"move\": {\"cost\": 0}, \"rotate\": {\"cost\": 0}, \"jump\": {\"cost\": 3}, \"shield\": {\"cost\": 0}, \"drop\": {\"cost\": 1}, \"use\": {\"cost\": 0}, \"attack\": {\"cost\": 5, \"damage\": 5}}, \"map\": {\"layout\": [[\"base\", \"wild_1\", \"base\"], [\"wild_2\", \"center\", \"wild_2\"], [\"base\", \"wild_1\", \"base\"]], \"room\": {\"width\": 25, \"height\": 25, \"num_agents\": 5, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 15, \"wall\": 10}}, \"wild_1\": {\"width\": 10, \"height\": 15, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 1, \"converter\": 1, \"generator\": 5, \"wall\": 5}}, \"wild_2\": {\"width\": 15, \"height\": 10, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 1, \"converter\": 1, \"generator\": 5, \"wall\": 5}}, \"center\": {\"width\": 10, \"height\": 10, \"border\": 0, \"objects\": {\"agent\": 0, \"altar\": 2, \"converter\": 5, \"generator\": 10, \"wall\": 20}}, \"base\": {\"width\": 15, \"height\": 15, \"border\": 1, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 5, \"wall\": 5}}}}}",
+    "agent_cfg": "{\"_target_\": \"agent.metta_agent.MettaAgent\", \"observation_encoders\": {\"grid_obs\": {\"feature_names\": [], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 512, \"layers\": 4}, \"global_vars\": {\"feature_names\": [], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}, \"last_action\": {\"feature_names\": [\"last_action_id\", \"last_action_val\"], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}, \"last_reward\": {\"feature_names\": [\"last_reward\"], \"normalize_features\": true, \"label_dim\": 4, \"output_dim\": 8, \"layers\": 2}}, \"fc\": {\"layers\": 3, \"output_dim\": 512}, \"decoder\": {\"_target_\": \"agent.decoder.Decoder\"}, \"core\": {\"_target_\": \"sample_factory.model.core.ModelCoreRNN\", \"rnn_type\": \"gru\", \"rnn_num_layers\": 1, \"rnn_size\": 512}}"
+  },
+  "git_hash": "6e97daa70c767bdbd708d90489bf1e623b94ae16",
+  "git_repo_name": "https://github.com/daveey/metta.git",
+  "wandb_unique_id": "p2.metta.6_20240610_092631_757116"
+}

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd636514e9de0f199d5517ebab91434743aca4e2954360a3a36ee313e54bc317
+size 1603862