Download Simulation/GelSight_mini/roughness_classify/smooth/33.hdf5 from FVL-Lab/OpenViTac: direct link, hf CLI and curl.
- Browser
- Download file 110 MB
-
https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/33.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/roughness_classify/smooth/33.hdf5
-
curl -L -o 33.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/33.hdf5
110 MB
- Xet hash:
- 3ae24e2fd9b3c9b9997c8331466688dd11b71cde5557860093b49a7827d072ef
- Size of remote file:
- 110 MB
- SHA256:
- 54c8a240679a6ad592e2c1c61d99636b3182bf8ecac0a8c4c5dc109c37e675c2
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.