Download Simulation/GelSight_mini/roughness_classify/smooth/9.hdf5 from FVL-Lab/OpenViTac: direct link, hf CLI and curl.
- Browser
- Download file 107 MB
-
https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/9.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/roughness_classify/smooth/9.hdf5
-
curl -L -o 9.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/9.hdf5
107 MB
- Xet hash:
- c46616977e0a85d6453610bb478b3225ed4aa10fae04995a877b8f6224cc5830
- Size of remote file:
- 107 MB
- SHA256:
- c27eb7883266655b3677b459b1429279d5175ede106e459a0f11c494b3a74f44
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.