Datasets:
Download cleaned/data.parquet from failed09/bashkir-wikipedia-parallel: direct link, hf CLI and curl.
- Browser
- Download file 15.8 MB
-
https://huggingface.co/datasets/failed09/bashkir-wikipedia-parallel/resolve/main/cleaned/data.parquet
- Command line
-
hf download hf://datasets/failed09/bashkir-wikipedia-parallel/cleaned/data.parquet
-
curl -L -o data.parquet https://huggingface.co/datasets/failed09/bashkir-wikipedia-parallel/resolve/main/cleaned/data.parquet
15.8 MB
- Xet hash:
- c1d77f867499c8722b751ada321f38618382bbbe3be1e7aef112ca4a0740c0dd
- Size of remote file:
- 15.8 MB
- SHA256:
- dbece47e97fa44fccba19d31ddd574cd069e0d843755c889ed757fe4e0dd0e62
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.