Download Simulation/GelSight_mini/roughness_classify/smooth/37.hdf5 from FVL-Lab/OpenViTac: direct link, hf CLI and curl.
- Browser
- Download file 108 MB
-
https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/37.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/roughness_classify/smooth/37.hdf5
-
curl -L -o 37.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/37.hdf5
108 MB
- Xet hash:
- 4ac33863959a65ed8eea395ffbf3fac15cba01faa8fcd2cb33d4f52f2ec916d0
- Size of remote file:
- 108 MB
- SHA256:
- cf96fad9d52c04a24a165c17d59af415ff23e0a77b8b4bf2f33fcd647ddaabee
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.