Upload README.md with huggingface_hub

README.md

@@ -1,62 +1,79 @@
 ---
-datasets: shraavb/pick_place_cube
-library_name: lerobot
 license: apache-2.0
-model_name: act
-pipeline_tag: robotics
 tags:
-- lerobot
 - robotics
+- lerobot
 - act
+- imitation-learning
+- so101
+- manipulation
 ---
 
-# Model Card for act
-
-<!-- Provide a quick summary of what the model is/does. -->
-
+# ACT Policy for Pick and Place Cube Task
 
-[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.
+This model was trained using [LeRobot](https://github.com/huggingface/lerobot) to perform a pick-and-place task with an SO-101 robot arm.
 
+## Task Description
 
-This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).
-See the full documentation at [LeRobot Docs](https://huggingface.co/docs/lerobot/index).
-
----
-
-## How to Get Started with the Model
+**Task:** Pick up a cube and place it in a container
 
-For a complete walkthrough, see the [training guide](https://huggingface.co/docs/lerobot/il_robots#train-a-policy).
-Below is the short version on how to train and run inference/eval:
+The robot learns to:
+1. Locate the cube visually
+2. Move the gripper to the cube
+3. Grasp the cube
+4. Transport it to the target container
+5. Release the cube into the container
 
-### Train from scratch
+## Model Details
 
-```bash
-lerobot-train \
-  --dataset.repo_id=${HF_USER}/<dataset> \
-  --policy.type=act \
-  --output_dir=outputs/train/<desired_policy_repo_id> \
-  --job_name=lerobot_training \
-  --policy.device=cuda \
-  --policy.repo_id=${HF_USER}/<desired_policy_repo_id>
-  --wandb.enable=true
+- **Policy Type:** ACT (Action Chunking with Transformers)
+- **Robot:** SO-101 Follower Arm (6-DOF with gripper)
+- **Camera:** Single front-facing RGB camera (640x480 @ 30fps)
+- **Training Steps:** 100,000
+- **Training Episodes:** 51 demonstrations
+- **Framework:** LeRobot
+
+## Training Configuration
+
+```yaml
+policy:
+  type: act
+  chunk_size: 100
+  n_action_steps: 100
+  dim_model: 512
+  n_heads: 8
+  n_encoder_layers: 4
+  n_decoder_layers: 1
+  vision_backbone: resnet18
+  use_vae: true
 ```
 
-_Writes checkpoints to `outputs/train/<desired_policy_repo_id>/checkpoints/`._
+## Usage
 
-### Evaluate the policy/run inference
+### Evaluation on Robot
 
 ```bash
 lerobot-record \
-  --robot.type=so100_follower \
-  --dataset.repo_id=<hf_user>/eval_<dataset> \
-  --policy.path=<hf_user>/<desired_policy_repo_id> \
-  --episodes=10
+  --robot.type=so101_follower \
+  --robot.port=/dev/ttyACM0 \
+  --robot.cameras='{ front: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 30} }' \
+  --dataset.repo_id=YOUR_USERNAME/eval_pick_place_cube \
+  --dataset.single_task="Pick up the cube and place it in the container" \
+  --policy.path=shraavb/act_pick_place_cube
 ```
 
-Prefix the dataset repo with **eval\_** and supply `--policy.path` pointing to a local or hub checkpoint.
+## Training Dataset
 
----
+- **Dataset:** [shraavb/pick_place_cube](https://huggingface.co/datasets/shraavb/pick_place_cube)
+- **Episodes:** 51
+- **Frames:** ~45,000
 
-## Model Details
+## Hardware Requirements
+
+- SO-101 robot arm with Feetech STS3215 servos
+- USB camera (webcam or similar)
+- Computer with USB ports for robot and camera
+
+## Acknowledgments
 
-- **License:** apache-2.0
+Trained following the [Seeed Studio LeRobot tutorial](https://wiki.seeedstudio.com/lerobot_so100m/).