Download Simulation/GelSight_mini/roughness_classify/smooth/36.hdf5 from FVL-Lab/OpenViTac: direct link, hf CLI and curl.
- Browser
- Download file 111 MB
-
https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/36.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/roughness_classify/smooth/36.hdf5
-
curl -L -o 36.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/36.hdf5
111 MB
- Xet hash:
- a39251c7ba37fcd50885e0cc6fed35640d709a386cd334904f5303152d24ae35
- Size of remote file:
- 111 MB
- SHA256:
- a7d74e0d8ca6ab3a8ed0cb44ceb9e7e73ea9e50a95b454d65ce77801db4a6f87
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.