Datasets:
Download data/test/CoffeeScript/part.0.parquet from andstor/the_pile_github: direct link, hf CLI and curl.
- Browser
- Download file 258 kB
-
https://huggingface.co/datasets/andstor/the_pile_github/resolve/main/data/test/CoffeeScript/part.0.parquet
- Command line
-
hf download hf://datasets/andstor/the_pile_github/data/test/CoffeeScript/part.0.parquet
-
curl -L -o part.0.parquet https://huggingface.co/datasets/andstor/the_pile_github/resolve/main/data/test/CoffeeScript/part.0.parquet
258 kB
- Xet hash:
- 1b2d72835cad15a7df1b966ba4a7ba136c8c9f20b13fb42b908762f81f263524
- Size of remote file:
- 258 kB
- SHA256:
- fe4fb6c8feb7500387d01c756bd6cf2281bf88b407e02cad7f96e35785ee06f8
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.