karpathy / covid-sanity

Aspires to help the influx of bioRxiv / medRxiv papers on COVID-19

Home Page:http://biomed-sanity.com/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Search Terms Examples

starkfree opened this issue · comments

I might have just overlooked it, but do you provide an example anywhere of how a person might customize the search terms being used? On quick glance, it seems like these would go in a search.json file, but there is currently none provided in the repo, and I didn't see a sample of how the file would be formatted.

Could you perhaps provide the search.json file being used currently to filter for COVID-19 related terms as a starting point for customization?

Thanks!

search.json is computed automatically in run.py as a result of the scikit-learn tokenizer. There is no way to manually add search terms right now. In principle there could be a manual list. What is an example of what you'd like to add?

Apologies, I just now realized the papers are being compiled by biorxiv at this URL
https://connect.biorxiv.org/relate/collection_json.php?grp=181

I assumed you were filtering them after the fact. Sorry for the confusion.