Download tokenized_dataset/train.parquet from nlpscu/Tokenized_Multi_Code-IR: direct link, hf CLI and curl.
- Browser
- Download file 4.6 GB
-
https://huggingface.co/datasets/nlpscu/Tokenized_Multi_Code-IR/resolve/main/tokenized_dataset/train.parquet
- Command line
-
hf download hf://datasets/nlpscu/Tokenized_Multi_Code-IR/tokenized_dataset/train.parquet
-
curl -L -o train.parquet https://huggingface.co/datasets/nlpscu/Tokenized_Multi_Code-IR/resolve/main/tokenized_dataset/train.parquet
4.6 GB
- Xet hash:
- 2ca5ea0549802491a7ab0295983c3a39f1983f12c46abea9832ed54d28c3c03f
- Size of remote file:
- 4.6 GB
- SHA256:
- b11b25e97eb3c3b9cfa6feb778898f786a8af4236a19b69a9332688064bf43cb
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.