A dockerized semantic similarity API for ranking Searches given a Query

Used 2 Rankers for ranking candidate strings given a search query:

1. Angular Similarity between USEQA embedded strings
2. BM25+

The API is a flask web app taking HTML form input

Deploying on local machine

1. Clone the repository
2. Set terminal working directory as Cloned repository

Build instructions

sudo docker build -t my_docker_flask:latest .

Run instructions

The app is binded to port 5000

sudo docker run -p 5000:5000 my_docker_flask:latest 

Check if the container is running

sudo docker ps

Working command

Go to the below page & run the app!

http://0.0.0.0:5000/get_response_form

In the candidates string box enter all the candidate strings separated by commas (that's how its being split & preprocessed).

eg: hello, i hate spinach, christmas is in december, iphone 12 will launch this year,.....