jellyho/bigym-ogbench-20hz

BiGym, state observations, OGBench layout, uniform 20 Hz Every BiGym task's demonstrations replayed once and stored in OGBench's on-disk layout: one row per STATE, so a trajectory of T transitions is T+1 rows, terminals marks the last row, and the final row's action is a pad that ogbench.utils.load_dataset drops. array meaning observations robot proprioception (63 or 66) followed by object state (3-53) actions joint position targets rescaled to [-1, 1] by… See the full description on the dataset page: https://huggingface.co/datasets/jellyho/bigym-ogbench-20hz.

<iframe src="https://silytics.si/embed/item/cc0275a7-d927-4f13-905c-06ae11c12768" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.