Datasets:
Tasks:
Text Generation
Formats:
parquet
Sub-tasks:
language-modeling
Languages:
Danish
Size:
1M - 10M
License:
Add Danish github repositories
#31
by
KennethEnevoldsen
- opened
There must be a ton of Danish github repositories out there. I could imagine that it would include a decent number of tokens.
The stack is probably a good place to start:
https://huggingface.co/datasets/bigcode/the-stack/viewer
Not sure if it contains Danish but searching for æøå would be a decent quick way to check.
Probably reasonable to follow up with a language clf
The stack contains ~500M docs assuming 0.1-0.01% is Danish and a doc is about 100-1000 tokens (no idea is that is true), then we get 10-500M tokens