Upload policy weights, train config and readme

README.md

@@ -2,20 +2,28 @@
 datasets: maskjp/piper_pick_f15
 library_name: lerobot
 license: apache-2.0
-model_name: act
+model_name: pi05
 pipeline_tag: robotics
 tags:
 - robotics
-- act
 - lerobot
+- pi05
 ---
 
-# Model Card for act
+# Model Card for pi05
 
 <!-- Provide a quick summary of what the model is/does. -->
 
 
-[Action Chunking with Transformers (ACT)](https://huggingface.co/papers/2304.13705) is an imitation-learning method that predicts short action chunks instead of single steps. It learns from teleoperated data and often achieves high success rates.
+**π₀.₅ (Pi05) Policy**
+
+π₀.₅ is a Vision-Language-Action model with open-world generalization, from Physical Intelligence. The LeRobot implementation is adapted from their open source OpenPI repository.
+
+**Model Overview**
+
+π₀.₅ represents a significant evolution from π₀, developed by Physical Intelligence to address a big challenge in robotics: open-world generalization. While robots can perform impressive tasks in controlled environments, π₀.₅ is designed to generalize to entirely new environments and situations that were never seen during training.
+
+For more details, see the [Physical Intelligence π₀.₅ blog post](https://www.physicalintelligence.company/blog/pi05).
 
 
 This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).

config.json

@@ -1,5 +1,5 @@
 {
-    "type": "act",
+    "type": "pi05",
     "n_obs_steps": 1,
     "input_features": {
         "observation.state": {
@@ -49,31 +49,60 @@
     "private": null,
     "tags": null,
     "license": null,
-    "pretrained_path": null,
-    "chunk_size": 100,
-    "n_action_steps": 100,
+    "pretrained_path": "lerobot/pi05_base",
+    "paligemma_variant": "gemma_2b",
+    "action_expert_variant": "gemma_300m",
+    "dtype": "bfloat16",
+    "chunk_size": 50,
+    "n_action_steps": 50,
+    "max_state_dim": 32,
+    "max_action_dim": 32,
+    "num_inference_steps": 10,
+    "time_sampling_beta_alpha": 1.5,
+    "time_sampling_beta_beta": 1.0,
+    "time_sampling_scale": 0.999,
+    "time_sampling_offset": 0.001,
+    "min_period": 0.004,
+    "max_period": 4.0,
+    "use_relative_actions": false,
+    "relative_exclude_joints": [
+        "gripper"
+    ],
+    "action_feature_names": [
+        "joint1.pos",
+        "joint2.pos",
+        "joint3.pos",
+        "joint4.pos",
+        "joint5.pos",
+        "joint6.pos",
+        "gripper.pos"
+    ],
+    "rtc_config": null,
+    "image_resolution": [
+        224,
+        224
+    ],
+    "empty_cameras": 0,
+    "tokenizer_max_length": 200,
     "normalization_mapping": {
-        "VISUAL": "MEAN_STD",
-        "STATE": "MEAN_STD",
-        "ACTION": "MEAN_STD"
+        "VISUAL": "IDENTITY",
+        "STATE": "QUANTILES",
+        "ACTION": "QUANTILES"
     },
-    "vision_backbone": "resnet18",
-    "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
-    "replace_final_stride_with_dilation": false,
-    "pre_norm": false,
-    "dim_model": 512,
-    "n_heads": 8,
-    "dim_feedforward": 3200,
-    "feedforward_activation": "relu",
-    "n_encoder_layers": 4,
-    "n_decoder_layers": 1,
-    "use_vae": true,
-    "latent_dim": 32,
-    "n_vae_encoder_layers": 4,
-    "temporal_ensemble_coeff": null,
-    "dropout": 0.1,
-    "kl_weight": 10.0,
-    "optimizer_lr": 1e-05,
-    "optimizer_weight_decay": 0.0001,
-    "optimizer_lr_backbone": 1e-05
+    "gradient_checkpointing": true,
+    "compile_model": true,
+    "compile_mode": "max-autotune",
+    "freeze_vision_encoder": false,
+    "train_expert_only": false,
+    "optimizer_lr": 2.5e-05,
+    "optimizer_betas": [
+        0.9,
+        0.95
+    ],
+    "optimizer_eps": 1e-08,
+    "optimizer_weight_decay": 0.01,
+    "optimizer_grad_clip_norm": 1.0,
+    "scheduler_warmup_steps": 1000,
+    "scheduler_decay_steps": 30000,
+    "scheduler_decay_lr": 2.5e-06
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:879d3787c27f32056130204e27cd5bc75ae7290818839ed30e30d745a08685bb
-size 206707932
+oid sha256:99a6cd587ce06040f7e7cb03fc4f504d8a04ed08593ca035b539d785ee6508e2
+size 9354050752

train_config.json

@@ -81,7 +81,7 @@
     },
     "env": null,
     "policy": {
-        "type": "act",
+        "type": "pi05",
         "n_obs_steps": 1,
         "input_features": {
             "observation.state": {
@@ -131,42 +131,71 @@
         "private": null,
         "tags": null,
         "license": null,
-        "pretrained_path": null,
-        "chunk_size": 100,
-        "n_action_steps": 100,
+        "pretrained_path": "lerobot/pi05_base",
+        "paligemma_variant": "gemma_2b",
+        "action_expert_variant": "gemma_300m",
+        "dtype": "bfloat16",
+        "chunk_size": 50,
+        "n_action_steps": 50,
+        "max_state_dim": 32,
+        "max_action_dim": 32,
+        "num_inference_steps": 10,
+        "time_sampling_beta_alpha": 1.5,
+        "time_sampling_beta_beta": 1.0,
+        "time_sampling_scale": 0.999,
+        "time_sampling_offset": 0.001,
+        "min_period": 0.004,
+        "max_period": 4.0,
+        "use_relative_actions": false,
+        "relative_exclude_joints": [
+            "gripper"
+        ],
+        "action_feature_names": [
+            "joint1.pos",
+            "joint2.pos",
+            "joint3.pos",
+            "joint4.pos",
+            "joint5.pos",
+            "joint6.pos",
+            "gripper.pos"
+        ],
+        "rtc_config": null,
+        "image_resolution": [
+            224,
+            224
+        ],
+        "empty_cameras": 0,
+        "tokenizer_max_length": 200,
         "normalization_mapping": {
-            "VISUAL": "MEAN_STD",
-            "STATE": "MEAN_STD",
-            "ACTION": "MEAN_STD"
+            "VISUAL": "IDENTITY",
+            "STATE": "QUANTILES",
+            "ACTION": "QUANTILES"
         },
-        "vision_backbone": "resnet18",
-        "pretrained_backbone_weights": "ResNet18_Weights.IMAGENET1K_V1",
-        "replace_final_stride_with_dilation": false,
-        "pre_norm": false,
-        "dim_model": 512,
-        "n_heads": 8,
-        "dim_feedforward": 3200,
-        "feedforward_activation": "relu",
-        "n_encoder_layers": 4,
-        "n_decoder_layers": 1,
-        "use_vae": true,
-        "latent_dim": 32,
-        "n_vae_encoder_layers": 4,
-        "temporal_ensemble_coeff": null,
-        "dropout": 0.1,
-        "kl_weight": 10.0,
-        "optimizer_lr": 1e-05,
-        "optimizer_weight_decay": 0.0001,
-        "optimizer_lr_backbone": 1e-05
+        "gradient_checkpointing": true,
+        "compile_model": true,
+        "compile_mode": "max-autotune",
+        "freeze_vision_encoder": false,
+        "train_expert_only": false,
+        "optimizer_lr": 2.5e-05,
+        "optimizer_betas": [
+            0.9,
+            0.95
+        ],
+        "optimizer_eps": 1e-08,
+        "optimizer_weight_decay": 0.01,
+        "optimizer_grad_clip_norm": 1.0,
+        "scheduler_warmup_steps": 1000,
+        "scheduler_decay_steps": 30000,
+        "scheduler_decay_lr": 2.5e-06
     },
-    "output_dir": "outputs/train/piper_pick_f15",
+    "output_dir": "outputs/piper_pick_f15",
     "job_name": "piper_pick_f15",
     "resume": false,
     "seed": 1000,
     "cudnn_deterministic": false,
     "num_workers": 4,
-    "batch_size": 8,
-    "steps": 100000,
+    "batch_size": 24,
+    "steps": 3000,
     "eval_freq": 20000,
     "log_freq": 200,
     "tolerance_s": 0.0001,
@@ -175,16 +204,22 @@
     "use_policy_training_preset": true,
     "optimizer": {
         "type": "adamw",
-        "lr": 1e-05,
-        "weight_decay": 0.0001,
-        "grad_clip_norm": 10.0,
+        "lr": 2.5e-05,
+        "weight_decay": 0.01,
+        "grad_clip_norm": 1.0,
         "betas": [
             0.9,
-            0.999
+            0.95
         ],
         "eps": 1e-08
     },
-    "scheduler": null,
+    "scheduler": {
+        "type": "cosine_decay_with_warmup",
+        "num_warmup_steps": 1000,
+        "num_decay_steps": 30000,
+        "peak_lr": 2.5e-05,
+        "decay_lr": 2.5e-06
+    },
     "eval": {
         "n_episodes": 50,
         "batch_size": 50,
@@ -196,7 +231,7 @@
         "project": "lerobot",
         "entity": null,
         "notes": null,
-        "run_id": "8gbwhr1i",
+        "run_id": "fysn8jv9",
         "mode": null,
         "add_tags": true
     },