Downloading Files Individually Different Than Cloning Repo

#11
by bgenchel - opened

When I download file by file by clicking on the download icon on the files and version tab, I can open the files with duckdb and pandas. When I clone the repo with git clone https://huggingface.co/datasets/PleIAs/YouTube-Commons and I try to read the files, I get this:

duckdb.duckdb.InvalidInputException: Invalid Input Error: No magic bytes found at end of file 'cctube_0.parquet'

(and something similar with pandas as well). What's going on?

Sign up or log in to comment