Download Simulation/GelSight_mini/roughness_classify/rough/29.hdf5 from FVL-Lab/OpenViTac: direct link, hf CLI and curl.
- Browser
- Download file 121 MB
-
https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/rough/29.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/roughness_classify/rough/29.hdf5
-
curl -L -o 29.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/rough/29.hdf5
121 MB
- Xet hash:
- f75ad202a2739cc9299264abfbec5858d4c2686b60ef9657641c02b266c6d7f2
- Size of remote file:
- 121 MB
- SHA256:
- 13ea276eeb27d0f4a5ba0a9a38f94c32b2142e379b96d7f2c2609b5235c7a2da
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.