How to reproduce the corpus?
#3
by
xianf
- opened
It said "You can collect those files referring to this metadata and reproduce RefineCode!" in the README, but how can we achieve that? Using wget to download the exact url get a webpage and I need to extract the content. Or do I need to download the full Stack-v2 dataset and extract from the original dataset?