This repository contains a Python Flask app that allows users to search for subfeatures based on a text query. The search is based on cosine similarity between the query and the descriptions of subfeatures in a provided dataset, the embeddings are generated by using Eden AI API.
- Python 3.x
- Flask
- pandas
- numpy
- requests
Start the Flask app by running app.py in the terminal:
python app.py
Open a web browser and navigate to http://localhost:5000/search?description=<query>
Replace <query> with your desired search query.
The app will return a JSON object containing the top 5 subfeatures with the highest cosine similarity to the query.