Vector Database
Getting Vectors
Start by running the python vector server
# cd python
# source venv/bin/activate
# pip install -r requirements.txt
# python app.py
This will start a server, please note that no models will of been cached. On the first request the model will be cached so will take some time.
Example request
# curl -XPOST -d '["test", "test"]' -H "Content-Type: application/json" http://127.0.0.1:5000/doc/embedding/msmarco-roberta-base-ance-fristp
Example response
[[0.9499503970146179,0.8039794564247131,0.022799743339419365,...],[0.9499503970146179,0.8039794564247131,0.022799743339419365,...]]
Setting up elasticsearch
Requires version 7.12+
Install elastiknn plugin from elastiknn.com
# ./bin/elasticsearch-plugin install --batch https://github.com/alexklibisz/elastiknn/releases/download/7.12.0.0/elastiknn-7.12.0.0.zip