Download .gitattributes from pdfqa/pdfQA-Benchmark: direct link, hf CLI and curl.
- Browser
- Download file 499 Bytes
-
https://huggingface.co/datasets/pdfqa/pdfQA-Benchmark/resolve/main/.gitattributes
- Command line
-
hf download hf://datasets/pdfqa/pdfQA-Benchmark/.gitattributes
-
curl -L -o .gitattributes https://huggingface.co/datasets/pdfqa/pdfQA-Benchmark/resolve/main/.gitattributes
499 Bytes
| # Big/binary formats -> LFS | |
| *.pdf filter=lfs diff=lfs merge=lfs -text | |
| *.xlsx filter=lfs diff=lfs merge=lfs -text | |
| *.xls filter=lfs diff=lfs merge=lfs -text | |
| *.epub filter=lfs diff=lfs merge=lfs -text | |
| # These often exceed HF's 10 MiB limit -> LFS | |
| *.csv filter=lfs diff=lfs merge=lfs -text | |
| *.htm filter=lfs diff=lfs merge=lfs -text | |
| *.html filter=lfs diff=lfs merge=lfs -text | |
| # "binary-ish" txt -> LFS | |
| *.txt filter=lfs diff=lfs merge=lfs -text | |
| # Normal text | |
| *.tex text eol=lf | |
| *.json text eol=lf | |