- Run
pip3 -r requirements.txt
- Download PubMed XML files from this URL and place them into
data
- Run
python3 embedding.py
. The embedding along with the documents are serialized (this may take a long time) - Run HTTP endpoint using
flask -A server run
- Access zipdoc as http://127.0.0.1:5000
macemoth/HackZurich2022
A medical search engine with vector search implemented at the Hack Zurich 2022 within 33 hours
Python