Upload folder using huggingface_hub

README.md

@@ -1,29 +1,107 @@
 ---
 library_name: physicalai
 license: apache-2.0
-model_name: act
+model_name: ACT
 pipeline_tag: robotics
 tags:
 - act
+- executorch
+- onnx
+- openvino
+- physical-ai-studio
 - physicalai
 - robotics
 - torch
+- vision-language-action
 ---
 
-# Model Card for act
+# Model Card for ACT
 
-[Action Chunking with Transformers (ACT)](https://huggingface.co/papers/2304.13705) is an imitation-learning policy that predicts short action chunks from robot state and visual observations.
+[Action Chunking with Transformers (ACT)](https://huggingface.co/papers/2304.13705) is an imitation-learning policy
+that predicts short action chunks from robot state and visual observations. The robot can execute those chunks as a
+sequence of real-world movements.
 
 This model was trained and exported with Physical AI Studio. It can be loaded from this directory with the root
 `manifest.json`, or from backend-specific manifests under `exports/<backend>/manifest.json`.
 
+## Model Details
+
+- **Policy:** act
+- **Runtime library:** `physicalai`
+- **Generated by:** Physical AI Studio
+- **Root manifest:** `manifest.json`
+- **Backend manifests:** `exports/<backend>/manifest.json`
+
+## Intended Use
+
+Use this model for robot imitation-learning inference in setups matching the training dataset, robot embodiment,
+camera viewpoints, and task instructions. Validate behavior in simulation or a safe test cell before running on hardware.
+
+## Dataset
+
+This model was trained from the Physical AI Studio dataset named **Dice cleanup**.
+
 ## Available Exports
 
 - **torch:** `exports/torch/act.pt`
+- **executorch:** `exports/executorch/act.pte`
+- **onnx:** `exports/onnx/act.onnx`
+- **openvino:** `exports/openvino/act.xml`
+
+## Training Environment
+
+Environment: **So101**
+
+### Robots
+
+| Name | Type | Teleoperator | Calibration |
+| --- | --- | --- | --- |
+| SO101 Follower | SO101_Follower | SO101 Leader (SO101_Leader) | Included |
+
+### Cameras
+
+| Name | Driver | Hardware | Resolution | FPS |
+| --- | --- | --- | --- | --- |
+| Gripper | usb_camera | Innomaker-U20CAM-1080p-S1: Inno | 640x480 | 30 |
+| Overview | usb_camera | Innomaker-U20CAM-1080p-S1: Inno | 640x480 | 30 |
 
 ## I/O Specification
 
-### Inputs
+### executorch
+
+#### Inputs
+
+| Name | Type | Shape | Dtype |
+| --- | --- | --- | --- |
+| state | STATE | [6] | float32 |
+| images.gripper | VISUAL | [3, 480, 640] | float32 |
+| images.overview | VISUAL | [3, 480, 640] | float32 |
+
+#### Outputs
+
+| Name | Type | Shape | Dtype |
+| --- | --- | --- | --- |
+| action | ACTION | [100, 6] | float32 |
+
+### onnx
+
+#### Inputs
+
+| Name | Type | Shape | Dtype |
+| --- | --- | --- | --- |
+| state | STATE | [6] | float32 |
+| images.gripper | VISUAL | [3, 480, 640] | float32 |
+| images.overview | VISUAL | [3, 480, 640] | float32 |
+
+#### Outputs
+
+| Name | Type | Shape | Dtype |
+| --- | --- | --- | --- |
+| action | ACTION | [100, 6] | float32 |
+
+### openvino
+
+#### Inputs
 
 | Name | Type | Shape | Dtype |
 | --- | --- | --- | --- |
@@ -31,23 +109,88 @@ This model was trained and exported with Physical AI Studio. It can be loaded fr
 | images.gripper | VISUAL | [3, 480, 640] | float32 |
 | images.overview | VISUAL | [3, 480, 640] | float32 |
 
-### Outputs
+#### Outputs
 
 | Name | Type | Shape | Dtype |
 | --- | --- | --- | --- |
 | action | ACTION | [100, 6] | float32 |
 
-## How To Run Inference
+### torch
 
-Use the `manifest.json` in this directory for the canonical Torch/checkpoint export, or use a backend-specific manifest
-from `exports/<backend>/manifest.json` for a specific runtime.
+#### Inputs
 
-## How To Retrain
+| Name | Type | Shape | Dtype |
+| --- | --- | --- | --- |
+| state | STATE | [6] | float32 |
+| images.gripper | VISUAL | [3, 480, 640] | float32 |
+| images.overview | VISUAL | [3, 480, 640] | float32 |
+
+#### Outputs
+
+| Name | Type | Shape | Dtype |
+| --- | --- | --- | --- |
+| action | ACTION | [100, 6] | float32 |
+
+## Running Inference
+
+### Installation
+
+```bash
+uv pip install physicalai numpy
+```
+
+The following example loads this model with the PhysicalAI inference API and builds a dummy observation matching the
+declared input specification:
+
+```python
+import numpy as np
+from physicalai.inference import InferenceModel
+
+MODEL_PATH = "path/to/model"
+model = InferenceModel.load(MODEL_PATH, device="CPU")
+
+observation = {
+    "state": np.random.rand(1, 6).astype(np.float32),
+    "images.gripper": np.random.rand(1, 3, 480, 640).astype(np.float32),
+    "images.overview": np.random.rand(1, 3, 480, 640).astype(np.float32),
+}
+
+chunk = model.predict_action_chunk(observation)
+```
+
+Set `MODEL_PATH` to this local model directory or to the Hugging Face repository id after upload.
+
+### Running A Robot Control Loop
+
+For a blocking control loop similar to PhysicalAI's `examples/runtime/sync_inference.py`, point the runtime at this
+model directory and provide your robot/camera configuration:
+
+```bash
+python examples/runtime/sync_inference.py \
+  --robot so101 \
+  --port /dev/ttyACM0 \
+  --calibration ./calibration.json \
+  --model path/to/model \
+  --camera overhead:uvc:/dev/video0 \
+  --task "pick up the object" \
+  --device CPU
+```
+
+## Training / Reproducing Training
 
 Import this model in Physical AI Studio and start a new training job using it as the base model. Studio will preserve
 the training lineage through the parent model relationship.
 
-## Limitations
+To reproduce behavior on your own hardware, match the exported I/O specification, robot type, camera viewpoints,
+control frequency, and calibration values from `environment.json` as closely as possible.
+
+## Evaluation
+
+No task-specific evaluation metrics were exported with this generated card. Add validation results, success rates, and
+hardware test conditions before publishing externally.
+
+## Limitations And Safety
 
-This generated model card summarizes package metadata and exported model I/O. Add task-specific evaluation results,
-dataset details, robot setup, and safety considerations before publishing externally.
+This generated model card summarizes package metadata, exported model I/O, and the sanitized training environment. Robot
+policies can behave unpredictably outside their training distribution. Use hardware limits, emergency stops, supervision,
+and staged validation before autonomous operation.

environment.json

@@ -0,0 +1,78 @@
+{
+  "format": "physical_ai_studio_environment",
+  "version": "1.0",
+  "name": "So101",
+  "robots": [
+    {
+      "name": "SO101 Follower",
+      "type": "SO101_Follower",
+      "calibration": {
+        "elbow_flex": {
+          "id": 3,
+          "drive_mode": 0,
+          "homing_offset": 1149,
+          "range_min": 851,
+          "range_max": 3074
+        },
+        "gripper": {
+          "id": 6,
+          "drive_mode": 0,
+          "homing_offset": 1088,
+          "range_min": 1938,
+          "range_max": 3416
+        },
+        "shoulder_lift": {
+          "id": 2,
+          "drive_mode": 0,
+          "homing_offset": 263,
+          "range_min": 821,
+          "range_max": 3195
+        },
+        "shoulder_pan": {
+          "id": 1,
+          "drive_mode": 0,
+          "homing_offset": 135,
+          "range_min": 732,
+          "range_max": 3454
+        },
+        "wrist_flex": {
+          "id": 4,
+          "drive_mode": 0,
+          "homing_offset": -1606,
+          "range_min": 860,
+          "range_max": 3188
+        },
+        "wrist_roll": {
+          "id": 5,
+          "drive_mode": 0,
+          "homing_offset": 612,
+          "range_min": 124,
+          "range_max": 3956
+        }
+      },
+      "teleoperator": {
+        "type": "robot",
+        "name": "SO101 Leader",
+        "robot_type": "SO101_Leader"
+      }
+    }
+  ],
+  "cameras": [
+    {
+      "name": "Gripper",
+      "driver": "usb_camera",
+      "hardware_name": "Innomaker-U20CAM-1080p-S1: Inno",
+      "width": 640,
+      "height": 480,
+      "fps": 30
+    },
+    {
+      "name": "Overview",
+      "driver": "usb_camera",
+      "hardware_name": "Innomaker-U20CAM-1080p-S1: Inno",
+      "width": 640,
+      "height": 480,
+      "fps": 30
+    }
+  ]
+}