Sign inGet started
Dataset · Hugging Face

chomeed/hmg-g1-drill-pnp-reward-train-success

HumanoidMimicGen G1 Drill Pnp: train / success Pick a drill from one table and place it on a second table. (HMG environment LMDrillPnP90.) The success episodes of the train split: 505 episodes, 584,042 frames at 50 Hz, one 480x640 ego-view RGB video each, with robot state, actions and per-frame task-success labels. LeRobot v3.0 format. How this split was made Derived from jellyho/hmg-g1-drill-pnp-reward by a group-level 90/10 train/test split. In the parent… See the full description on the dataset page: https://huggingface.co/datasets/chomeed/hmg-g1-drill-pnp-reward-train-success.

Author / organizationchomeed
License (as declared)apache-2.0
Downloads0 (as reported by Hugging Face)
Likes0
Created18 October 2026
Last updated at source18 October 2026
humanoidunitree-g1mujocolerobotreward-modelsuccess-detectionexpert-demonstrations
Embed this entry
<iframe src="https://silytics.si/embed/item/5a592a7c-81a6-4adb-9405-107716fdd88d" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Recent changes

  • 18 Oct 2026 Description: HumanoidMimicGen G1 Drill Pnp: train / success Pick a drill from one table and place it on a second table. (HMG environment LMDrillPnP90.) The success episodes of the train split: 505 episodes, 584,042 frames at 50 Hz, one 480x640 ego-view → HumanoidMimicGen G1 Drill Pnp: train / success Pick a drill from one table and place it on a second table. (HMG environment LMDrillPnP90.) The success episodes of the train split: 505 episodes, 584,042 frames at 50 Hz, one 480x640 ego-view
  • 18 Oct 2026 License: none → apache-2.0
  • 18 Oct 2026 Description: none → HumanoidMimicGen G1 Drill Pnp: train / success Pick a drill from one table and place it on a second table. (HMG environment LMDrillPnP90.) The success episodes of the train split: 505 episodes, 584,042 frames at 50 Hz, one 480x640 ego-view
  • 18 Oct 2026 Topics added task_categories:robotics, humanoid, unitree-g1, mujoco, lerobot, reward-model, success-detection, expert-demonstrations

Changes detected by Silytics between refreshes of the source.

More from chomeed

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.