ganatrask
/

act-panda-open-cabinet-right

Model card Files Files and versions

ACT Model for Panda-Open-Cabinet-Right Task

This model was trained using the Action Chunking Transformer (ACT) policy on the panda-open-cabinet-right task from NVIDIA's PhysicalAI-Robotics-Manipulation-SingleArm dataset.

Training Details

Steps: 200,000
Chunk size: 100
Layers: 4 encoder layers
Batch size: 32
Device: A100 GPU
AMP: Enabled

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support