Download Simulation/GelSight_mini/roughness_classify/smooth/4.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/4.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/roughness_classify/smooth/4.hdf5
-
curl -L -o 4.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/4.hdf5
107 MB
- Xet hash:
- 7671aaa5a89f9c70ffcd0734d27f7223d4887b5d40232551efd24f199cf9361b
- Size of remote file:
- 107 MB
- SHA256:
- c1f34f05eefbbfb70c8d73010cfdbfbc6dff5e5ee61095699900b1a1c8f7f97f
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.