Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,8 @@ 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
+sf_log.txt filter=lfs diff=lfs merge=lfs -text

.summary/0/events.out.tfevents.1719003583.a228d6d534a1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:720afb546337648336bd32ad625d97609062f9733f1dcf43eb351cf679864fc6
+size 319775646

.summary/0/events.out.tfevents.1719380402.a228d6d534a1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6a97bb9a0873756ff88949f65d1e58b6265ed5ee8920e63e2948e5eadb7330e
+size 40

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.v0.5.5
+```
+
+    
+## 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.v0.5.5
+```
+
+
+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.v0.5.5 --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 sha256:18533b294f488dd4e785e2f1fdfdaa691bf82b45410181ed7c818b9711dd9340
+size 2002553

a20_b4_40x40.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2002c7c68775b12235bd3b468a9851ffdaef79400091ceb367523b6ed621eb0a
+size 2248841

a20_r4_40x40.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e55450e0eea72105a88b3d6f1a2d9660b9b8d8beeb744862938f41f8be56250
+size 2359699

checkpoint_p0/best_000755114_12371787776_reward_0.948.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed92aa676c393a60db041b043616b7b05537533bb9d86b3e786abce51149d314
+size 39046778

checkpoint_p0/checkpoint_000980417_16063152128.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcca43dd239da730d227114fee3fda64832289d7fe39e09213947879b747a80b
+size 39048218

checkpoint_p0/checkpoint_000980730_16068280320.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7e099530d05fa7407ff36f910da7c2966e4a6aae7fe0f10ab149b1ee400a8ab
+size 39048218

config.json

@@ -0,0 +1,162 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "GDY-MettaGrid",
+  "experiment": "p2.dr6",
+  "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": "train_dir/baseline.v0.5.4/checkpoint_p0/checkpoint_000349863_5732155392.pth",
+  "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, \"hidden_features\": {\"grid_obs\": [\"agent:energy\", \"agent:hp\"]}, \"game\": {\"obs_width\": 11, \"obs_height\": 11, \"max_steps\": 5000, \"tile_size\": 16, \"num_agents\": 20, \"no_energy_steps\": 500, \"objects\": {\"agent\": {\"initial_energy\": [50, 250], \"max_energy\": [200, 250], \"max_inventory\": [3, 10], \"freeze_duration\": [5, 20], \"energy_reward\": false, \"hp\": 1, \"mortal\": false, \"upkeep\": {\"time\": 0, \"shield\": [0, 3]}, \"use_cost\": 0}, \"altar\": {\"hp\": [10, 50], \"cooldown\": [2, 5], \"use_cost\": [50, 250]}, \"converter\": {\"hp\": [10, 50], \"cooldown\": [2, 5], \"energy_output\": [50, 200], \"use_cost\": 0}, \"generator\": {\"hp\": [10, 50], \"cooldown\": [3, 10], \"initial_resources\": [5, 50], \"use_cost\": 0}, \"wall\": {\"density\": 0.01, \"hp\": [3, 20]}}, \"actions\": {\"move\": {\"cost\": [0, 2]}, \"rotate\": {\"cost\": [0, 1]}, \"jump\": {\"cost\": [1, 5]}, \"shield\": {\"cost\": [0, 2]}, \"drop\": {\"cost\": [0, 2]}, \"use\": {\"cost\": [0, 2]}, \"attack\": {\"cost\": [3, 10], \"damage\": [3, 10]}}, \"map\": {\"layout\": {\"rooms_x\": 2, \"rooms_y\": 2}, \"room\": {\"width\": 20, \"height\": 20, \"num_agents\": 5, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 10, \"wall\": 40}, \"border\": 1}}}, \"kinship\": {\"team_size\": [1, 5], \"team_reward\": [0.0, 1.0]}}",
+  "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}, \"kinship\": {\"feature_names\": [\"kinship\"], \"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_coeff=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.dr6 --init_checkpoint_path=train_dir/baseline.v0.5.4/checkpoint_p0/checkpoint_000349863_5732155392.pth --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, \"hidden_features\": {\"grid_obs\": [\"agent:energy\", \"agent:hp\"]}, \"game\": {\"obs_width\": 11, \"obs_height\": 11, \"max_steps\": 5000, \"tile_size\": 16, \"num_agents\": 20, \"no_energy_steps\": 500, \"objects\": {\"agent\": {\"initial_energy\": [50, 250], \"max_energy\": [200, 250], \"max_inventory\": [3, 10], \"freeze_duration\": [5, 20], \"energy_reward\": false, \"hp\": 1, \"mortal\": false, \"upkeep\": {\"time\": 0, \"shield\": [0, 3]}, \"use_cost\": 0}, \"altar\": {\"hp\": [10, 50], \"cooldown\": [2, 5], \"use_cost\": [50, 250]}, \"converter\": {\"hp\": [10, 50], \"cooldown\": [2, 5], \"energy_output\": [50, 200], \"use_cost\": 0}, \"generator\": {\"hp\": [10, 50], \"cooldown\": [3, 10], \"initial_resources\": [5, 50], \"use_cost\": 0}, \"wall\": {\"density\": 0.01, \"hp\": [3, 20]}}, \"actions\": {\"move\": {\"cost\": [0, 2]}, \"rotate\": {\"cost\": [0, 1]}, \"jump\": {\"cost\": [1, 5]}, \"shield\": {\"cost\": [0, 2]}, \"drop\": {\"cost\": [0, 2]}, \"use\": {\"cost\": [0, 2]}, \"attack\": {\"cost\": [3, 10], \"damage\": [3, 10]}}, \"map\": {\"layout\": {\"rooms_x\": 2, \"rooms_y\": 2}, \"room\": {\"width\": 20, \"height\": 20, \"num_agents\": 5, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 10, \"wall\": 40}, \"border\": 1}}}, \"kinship\": {\"team_size\": [1, 5], \"team_reward\": [0.0, 1.0]}} --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}, \"kinship\": {\"feature_names\": [\"kinship\"], \"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.dr6",
+    "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,
+    "init_checkpoint_path": "train_dir/baseline.v0.5.4/checkpoint_p0/checkpoint_000349863_5732155392.pth",
+    "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, \"hidden_features\": {\"grid_obs\": [\"agent:energy\", \"agent:hp\"]}, \"game\": {\"obs_width\": 11, \"obs_height\": 11, \"max_steps\": 5000, \"tile_size\": 16, \"num_agents\": 20, \"no_energy_steps\": 500, \"objects\": {\"agent\": {\"initial_energy\": [50, 250], \"max_energy\": [200, 250], \"max_inventory\": [3, 10], \"freeze_duration\": [5, 20], \"energy_reward\": false, \"hp\": 1, \"mortal\": false, \"upkeep\": {\"time\": 0, \"shield\": [0, 3]}, \"use_cost\": 0}, \"altar\": {\"hp\": [10, 50], \"cooldown\": [2, 5], \"use_cost\": [50, 250]}, \"converter\": {\"hp\": [10, 50], \"cooldown\": [2, 5], \"energy_output\": [50, 200], \"use_cost\": 0}, \"generator\": {\"hp\": [10, 50], \"cooldown\": [3, 10], \"initial_resources\": [5, 50], \"use_cost\": 0}, \"wall\": {\"density\": 0.01, \"hp\": [3, 20]}}, \"actions\": {\"move\": {\"cost\": [0, 2]}, \"rotate\": {\"cost\": [0, 1]}, \"jump\": {\"cost\": [1, 5]}, \"shield\": {\"cost\": [0, 2]}, \"drop\": {\"cost\": [0, 2]}, \"use\": {\"cost\": [0, 2]}, \"attack\": {\"cost\": [3, 10], \"damage\": [3, 10]}}, \"map\": {\"layout\": {\"rooms_x\": 2, \"rooms_y\": 2}, \"room\": {\"width\": 20, \"height\": 20, \"num_agents\": 5, \"objects\": {\"agent\": 5, \"altar\": 1, \"converter\": 3, \"generator\": 10, \"wall\": 40}, \"border\": 1}}}, \"kinship\": {\"team_size\": [1, 5], \"team_reward\": [0.0, 1.0]}}",
+    "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}, \"kinship\": {\"feature_names\": [\"kinship\"], \"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": "bd6dd5e8f31642adff63d12f41f4ddd3333616af",
+  "git_repo_name": "https://github.com/daveey/metta.git",
+  "wandb_unique_id": "p2.dr6_20240621_135940_570912"
+}

git.diff

@@ -0,0 +1,14 @@
+diff --git a/third_party/griddly b/third_party/griddly
+index 08f42c2..499412c 160000
+--- a/third_party/griddly
++++ b/third_party/griddly
+@@ -1 +1 @@
+-Subproject commit 08f42c2aea7a6d7e6bc50a672c748b47028cbf52
++Subproject commit 499412cb1eda60deb93d8f71a4d8896917a9a60c
+diff --git a/third_party/sample_factory b/third_party/sample_factory
+index 9d990ea..f9d0645 160000
+--- a/third_party/sample_factory
++++ b/third_party/sample_factory
+@@ -1 +1 @@
+-Subproject commit 9d990eaf3fd40c635febe5b3e67d5c7d3bb7a9cb
++Subproject commit f9d06453cc0e67ebc14d2e86c8d8d5c62e1389de

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2002c7c68775b12235bd3b468a9851ffdaef79400091ceb367523b6ed621eb0a
+size 2248841

sf_log.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8994115790c628ae58e09ffbf5c67d1135b94ac797854274c044f10ad1d7cd65
+size 33533435