Download Simulation/GelSight_mini/roughness_classify/smooth/2.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/2.hdf5
- Command line
-
hf download hf://datasets/FVL-Lab/OpenViTac/Simulation/GelSight_mini/roughness_classify/smooth/2.hdf5
-
curl -L -o 2.hdf5 https://huggingface.co/datasets/FVL-Lab/OpenViTac/resolve/main/Simulation/GelSight_mini/roughness_classify/smooth/2.hdf5
111 MB
- Xet hash:
- 6dbd84c8c9d4f111d7d779fe859fd5c42fd7654e43a36a06341f7e7b62a1e78c
- Size of remote file:
- 111 MB
- SHA256:
- 87a98b6b0f506d8860006c50f482d33328bb66656927bd640e06ac04b3874dc9
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.