PhysicalAI-Robotics-GR00T-X-Embodiment-Sim

Github Repo: Isaac GR00T N1

We provide a set of datasets used for post-training of GR00T N1. Each dataset is a collection of trajectories from different robot embodiments and tasks.

Cross-embodied bimanual manipulation: 9k trajectories

Humanoid robot tabletop manipulation: 240k trajectories

Humanoid robot tabletop manipulation - downsampled: 24k trajectories

Robot Arm Kitchen Manipulation: 72K trajectories

Unitree G1 Loco-Manipulation: 102 trajectories

Download the dataset from Huggingface

Users can download a specific subset of data by specifying the dataset name.

1. Option 1: with huggingface-cli

huggingface-cli download nvidia/PhysicalAI-Robotics-GR00T-X-Embodiment-Sim \
  --repo-type dataset \
  --include "gr1_arms_only.CanSort/**" \
  --local-dir $HOME/gr00t_dataset

Replace gr1_arms_only.CanSort/** with the dataset name you want to download.

2. Option 2: Github LFS

# Clone the repo without downloading files
git clone --filter=blob:none --no-checkout https://huggingface.co/datasets/nvidia/PhysicalAI-Robotics-GR00T-X-Embodiment-Sim
cd PhysicalAI-Robotics-GR00T-X-Embodiment-Sim

# Configure sparse-checkout for just your folder
git sparse-checkout init --cone
git sparse-checkout set gr1_arms_waist.CupToDrawer   << Select the specific dataset to download

Replace the "gr1_arms_waist.CupToDrawer" with the intended dataset folder