Datasets:
Download data/benchmark_split/train/benchmark_pre_30s.csv from YYYYUN/MERIT: direct link, hf CLI and curl.
- Browser
- Download file 30.7 MB
-
https://huggingface.co/datasets/YYYYUN/MERIT/resolve/main/data/benchmark_split/train/benchmark_pre_30s.csv
- Command line
-
hf download hf://datasets/YYYYUN/MERIT/data/benchmark_split/train/benchmark_pre_30s.csv
-
curl -L -o benchmark_pre_30s.csv https://huggingface.co/datasets/YYYYUN/MERIT/resolve/main/data/benchmark_split/train/benchmark_pre_30s.csv
30.7 MB
- Xet hash:
- d0a446d166ce94540f0cbcd0cf40c1f1b090a3d27c84b8b7ae79ead8fa41589a
- Size of remote file:
- 30.7 MB
- SHA256:
- 2a0fe0139cee41c09a911ec10ce7b9c692cc7c5df733e4da21a004ea925ff645
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.