Download Simulation/GelSight_mini/hardness_classify/soft/2.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/hardness_classify/soft/2.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/hardness_classify/soft/2.hdf5
-
curl -L -o 2.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/hardness_classify/soft/2.hdf5
108 MB
- Xet hash:
- 637af36ed33728be687b458fefd96b0ede5781b1709d9db66a4953fe5784a9f8
- Size of remote file:
- 108 MB
- SHA256:
- af7bdaef4f8cfd3ac30ed4eb79b605681ba9058df557beaa567fee810ae0d773
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.