https://github.com/xettrisomeman/movie-semantic-search

Search movies using natural language.
https://github.com/xettrisomeman/movie-semantic-search

cohere fastapi pinecone python semantic-search

Last synced: about 16 hours ago
JSON representation

Search movies using natural language.

• Host: GitHub
• URL: https://github.com/xettrisomeman/movie-semantic-search
• Owner: xettrisomeman
• License: mit
• Created: 2022-12-23T10:22:25.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2023-08-06T11:31:24.000Z (over 1 year ago)
• Last Synced: 2024-12-22T14:56:20.565Z (12 days ago)
• Topics: cohere, fastapi, pinecone, python, semantic-search
• Language: HTML
• Homepage: https://movie-semantic-search.vercel.app/
• Size: 1.57 MB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        


 





Natural Language Code For Film





Find movies in seconds



> Part of the project [Semantic Search AI Hackathon](https://lablab.ai/event/semantic-search-hackathon)

## Tools used

- [Cohere Ai](https://cohere.ai/)

- [PineCone](https://www.pinecone.io/)

### Running Locally

- git clone the repo

- ``pip install -r requirements.txt``

- create [pinecone](https://www.pinecone.io) and [cohere](https://cohere.ai/) account

- create ``.env`` file inside project directory and write

        ``PINECONE_API=``

       `` COHERE_API=``

- run ``python main.py``

### How we made it

Dataset: [Movies Daily Update Dataset](https://www.kaggle.com/datasets/akshaypawar7/millions-of-movies)

- The vector size is 4096(cohere-large). 

- Only 10k of 700k data being used. 

- No FineTuning!! No Preprocessing!!