Datasets:
Tasks:
Text Generation
Modalities:
Text
Sub-tasks:
language-modeling
Languages:
English
Size:
10M - 100M
License:
Download python/train/0000.parquet from shibing624/source_code: direct link, hf CLI and curl.
- Browser
- Download file 84.2 MB
-
https://huggingface.co/datasets/shibing624/source_code/resolve/refs%2Fconvert%2Fparquet/python/train/0000.parquet
- Command line
-
hf download hf://datasets/shibing624/source_code@refs/convert/parquet/python/train/0000.parquet
-
curl -L -o 0000.parquet https://huggingface.co/datasets/shibing624/source_code/resolve/refs%2Fconvert%2Fparquet/python/train/0000.parquet
84.2 MB
- Xet hash:
- 723cca708aa898ed69f3c43b3e79179723ad209011c7e6aeb2e824ae630a602f
- Size of remote file:
- 84.2 MB
- SHA256:
- d51766b442baba82a955b05a6f299c252bca2d9c232a55d418d98fa7d44a5054
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.