Upload with huggingface_hub

.summary/0/events.out.tfevents.1657653433.andrew-gpu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:748df02dd3566c234c1865c8baa5bc4dd5eb6d2e16d14baaf86c08b8e74313a4
+size 13908529

.summary/0/events.out.tfevents.1657748005.andrew-gpu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b87a967636a7230e0fa9cbedb470794924039215c06b5d9f3f7df06b546a695
+size 40

README.md

@@ -0,0 +1,11 @@
+---
+library_name: sample-factory
+tags:
+- deep-reinforcement-learning
+- reinforcement-learning
+- sample-factory
+---
+
+A(n) **APPO** model trained on the **quadrotor_multi** environment.
+This model was trained using Sample Factory 2.0: https://github.com/alex-petrenko/sample-factory
+    

cfg.json

@@ -0,0 +1,195 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "quadrotor_multi",
+  "experiment": "00_baseline_see_0",
+  "train_dir": "./train_dir/quad_single_baseline/baseline_",
+  "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": 100000000,
+  "num_workers": 36,
+  "num_envs_per_worker": 4,
+  "batch_size": 1024,
+  "num_batches_per_epoch": 1,
+  "num_epochs": 1,
+  "rollout": 128,
+  "recurrence": 1,
+  "shuffle_minibatches": true,
+  "gamma": 0.99,
+  "reward_scale": 1.0,
+  "reward_clip": 1000.0,
+  "value_bootstrap": false,
+  "normalize_returns": false,
+  "exploration_loss_coeff": 0.0,
+  "value_loss_coeff": 0.5,
+  "kl_loss_coeff": 0.0,
+  "exploration_loss": "entropy",
+  "gae_lambda": 1.0,
+  "ppo_clip_ratio": 0.1,
+  "ppo_clip_value": 5.0,
+  "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": 5.0,
+  "learning_rate": 0.0001,
+  "lr_schedule": "constant",
+  "lr_schedule_kl_threshold": 0.008,
+  "obs_subtract_mean": 0.0,
+  "obs_scale": 1.0,
+  "normalize_input": false,
+  "decorrelate_experience_max_seconds": 10,
+  "decorrelate_envs_on_one_worker": true,
+  "actor_worker_gpus": [],
+  "set_workers_cpu_affinity": true,
+  "force_envs_single_thread": true,
+  "default_niceness": 0,
+  "experiment_summaries_interval": 60,
+  "stats_avg": 100,
+  "train_for_env_steps": 10000000000,
+  "train_for_seconds": 10000000000,
+  "save_every_sec": 120,
+  "keep_checkpoints": 3,
+  "load_checkpoint_kind": "latest",
+  "save_milestones_sec": 10000,
+  "save_best_every_sec": 5,
+  "save_best_metric": "reward",
+  "save_best_after": 100000,
+  "benchmark": false,
+  "encoder_type": "mlp",
+  "encoder_subtype": "mlp_quads",
+  "encoder_custom": "quad_multi_encoder",
+  "encoder_extra_fc_layers": 0,
+  "hidden_size": 16,
+  "nonlinearity": "tanh",
+  "policy_initialization": "xavier_uniform",
+  "policy_init_gain": 1.0,
+  "actor_critic_share_weights": false,
+  "adaptive_stddev": false,
+  "initial_stddev": 1.0,
+  "use_rnn": false,
+  "rnn_type": "gru",
+  "rnn_num_layers": 1,
+  "env_gpu_actions": false,
+  "env_frameskip": 1,
+  "env_framestack": 4,
+  "pixel_format": "CHW",
+  "with_wandb": true,
+  "wandb_user": null,
+  "wandb_project": "sample_factory",
+  "wandb_group": null,
+  "wandb_job_type": "SF",
+  "wandb_tags": [
+    "sf2"
+  ],
+  "quads_discretize_actions": -1,
+  "quads_clip_input": false,
+  "quads_effort_reward": null,
+  "quads_episode_duration": 15.0,
+  "quads_num_agents": 1,
+  "quads_neighbor_hidden_size": 256,
+  "quads_neighbor_encoder_type": "no_encoder",
+  "quads_collision_reward": 5.0,
+  "quads_collision_obstacle_reward": 0.0,
+  "quads_settle_reward": 0.0,
+  "quads_settle": false,
+  "quads_vel_reward_out_range": 0.8,
+  "quads_settle_range_meters": 1.0,
+  "quads_collision_hitbox_radius": 2.0,
+  "quads_collision_falloff_radius": 4.0,
+  "quads_collision_smooth_max_penalty": 10.0,
+  "neighbor_obs_type": "none",
+  "quads_use_numba": true,
+  "quads_obstacle_mode": "no_obstacles",
+  "quads_obstacle_num": 0,
+  "quads_obstacle_type": "sphere",
+  "quads_obstacle_size": 0.0,
+  "quads_obstacle_traj": "gravity",
+  "quads_local_obs": -1,
+  "quads_local_coeff": 1.0,
+  "quads_local_metric": "dist",
+  "quads_view_mode": "local",
+  "quads_adaptive_env": false,
+  "quads_mode": "mix",
+  "quads_formation": "circle_horizontal",
+  "quads_formation_size": 0.0,
+  "room_dims": [
+    10,
+    10,
+    10
+  ],
+  "quads_obs_repr": "xyz_vxyz_R_omega",
+  "replay_buffer_sample_prob": 0.0,
+  "anneal_collision_steps": 0.0,
+  "quads_obstacle_obs_mode": "relative",
+  "quads_obstacle_hidden_size": 32,
+  "quads_collision_obst_smooth_max_penalty": 10.0,
+  "quads_obst_penalty_fall_off": 10.0,
+  "use_spectral_norm": false,
+  "command_line": "--env=quadrotor_multi --train_for_env_steps=1000000000 --algo=APPO --use_rnn=False --num_workers=36 --num_envs_per_worker=4 --learning_rate=0.0001 --ppo_clip_value=5.0 --recurrence=1 --nonlinearity=tanh --actor_critic_share_weights=False --policy_initialization=xavier_uniform --adaptive_stddev=False --with_vtrace=False --max_policy_lag=100000000 --hidden_size=256 --gae_lambda=1.00 --max_grad_norm=5.0 --exploration_loss_coeff=0.0 --rollout=128 --batch_size=1024 --quads_use_numba=True --quads_mode=mix --quads_episode_duration=15.0 --quads_formation_size=0.0 --encoder_custom=quad_multi_encoder --quads_collision_reward=5.0 --quads_neighbor_hidden_size=256 --neighbor_obs_type=pos_vel --quads_settle_reward=0.0 --quads_collision_hitbox_radius=2.0 --quads_collision_falloff_radius=4.0 --quads_local_obs=6 --quads_local_metric=dist --quads_local_coeff=1.0 --quads_num_agents=8 --quads_collision_reward=5.0 --quads_collision_smooth_max_penalty=10.0 --quads_neighbor_encoder_type=attention --replay_buffer_sample_prob=0.75 --anneal_collision_steps=300000000 --train_for_env_steps=10000000000 --hidden_size=16 --neighbor_obs_type=none --quads_local_obs=-1 --quads_num_agents=1 --replay_buffer_sample_prob=0.0 --anneal_collision_steps=0 --save_milestones_sec=10000 --quads_neighbor_encoder_type=no_encoder --serial_mode=False --with_wandb=True --wandb_tags sf2 --seed=0 --experiment=00_baseline_see_0 --train_dir=./train_dir/quad_single_baseline/baseline_",
+  "cli_args": {
+    "algo": "APPO",
+    "env": "quadrotor_multi",
+    "experiment": "00_baseline_see_0",
+    "train_dir": "./train_dir/quad_single_baseline/baseline_",
+    "seed": 0,
+    "serial_mode": false,
+    "max_policy_lag": 100000000,
+    "num_workers": 36,
+    "num_envs_per_worker": 4,
+    "batch_size": 1024,
+    "rollout": 128,
+    "recurrence": 1,
+    "exploration_loss_coeff": 0.0,
+    "gae_lambda": 1.0,
+    "ppo_clip_value": 5.0,
+    "with_vtrace": false,
+    "max_grad_norm": 5.0,
+    "learning_rate": 0.0001,
+    "train_for_env_steps": 10000000000,
+    "save_milestones_sec": 10000,
+    "encoder_custom": "quad_multi_encoder",
+    "hidden_size": 16,
+    "nonlinearity": "tanh",
+    "policy_initialization": "xavier_uniform",
+    "actor_critic_share_weights": false,
+    "adaptive_stddev": false,
+    "use_rnn": false,
+    "with_wandb": true,
+    "wandb_tags": [
+      "sf2"
+    ],
+    "quads_episode_duration": 15.0,
+    "quads_num_agents": 1,
+    "quads_neighbor_hidden_size": 256,
+    "quads_neighbor_encoder_type": "no_encoder",
+    "quads_collision_reward": 5.0,
+    "quads_settle_reward": 0.0,
+    "quads_collision_hitbox_radius": 2.0,
+    "quads_collision_falloff_radius": 4.0,
+    "quads_collision_smooth_max_penalty": 10.0,
+    "neighbor_obs_type": "none",
+    "quads_use_numba": true,
+    "quads_local_obs": -1,
+    "quads_local_coeff": 1.0,
+    "quads_local_metric": "dist",
+    "quads_mode": "mix",
+    "quads_formation_size": 0.0,
+    "replay_buffer_sample_prob": 0.0,
+    "anneal_collision_steps": 0.0
+  },
+  "git_hash": "e6745398a0413737776f3cd1681cf77173b05921",
+  "git_repo_name": "https://github.com/andrewzhang505/sample-factory.git",
+  "wandb_unique_id": "00_baseline_see_0_20220712_191712_050475"
+}

checkpoint_p0/best_000433823_444234752_reward_2.295.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0d9261c1f8dc571f09623c6abaca4ecc2597bf356e220db73958dd153f16a72
+size 45068

checkpoint_p0/checkpoint_000434271_444693504.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e08a4fe9a8f92717dc17078f72ba970e48ed9227b34ace3d96e0c273f4fe857
+size 45068

checkpoint_p0/checkpoint_000434837_445273088.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86d8e77863878e0a2aaab9b12c3ee4113160397bc5a0586e04d78190cabcb062
+size 45068

checkpoint_p0/checkpoint_000435337_445785088.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36b83a0ab91f0bb3b86414f188dcd44358049faea08c1b9469f437221a1c81be
+size 45068

git.diff

@@ -0,0 +1,14 @@
+diff --git a/sf_examples/swarm_rl_examples/runs/single_drone.py b/sf_examples/swarm_rl_examples/runs/single_drone.py
+index 03de717..4652986 100644
+--- a/sf_examples/swarm_rl_examples/runs/single_drone.py
++++ b/sf_examples/swarm_rl_examples/runs/single_drone.py
+@@ -10,7 +10,8 @@ _params = ParamGrid(
+ SMALL_MODEL_CLI = QUAD_BASELINE_CLI + (
+     " --train_for_env_steps=10000000000 --hidden_size=16 --neighbor_obs_type=none --quads_local_obs=-1 "
+     "--quads_num_agents=1 --replay_buffer_sample_prob=0.0 --anneal_collision_steps=0 --save_milestones_sec=10000 "
+-    "--quads_neighbor_encoder_type=no_encoder --serial_mode=False --with_wandb=True --wandb_tags sf2"
++    "--quads_neighbor_encoder_type=no_encoder --serial_mode=False --with_wandb=True --wandb_tags sf2 sync"
++    "--async_rl=False --reward_clip=10.0"
+ )
+ 
+ _experiment = Experiment(