A langchain movies demo in XetHub

We use langchain - index and openai to answer questions about Hollywood movies.

Requirements

pip install -r requirements.txt
export OPENAI_API_KEY=YOUR_OPENAI_API_KEY

Usage

Train

# Retrain from scratch
python src/train.py

# Insert a file  
python src/put.py --branch=temp --hackx=10 tests/data/fresh_prince.txt data

# Remove a file
python src/delete.py --branch=temp data/fresh_prince.txt

Run the app

gradio app.py

Development story

Simply forked Yonatan's langchain_demo Xet repo from XetHub and then I wanted to build a similar app for movie trivia.

I grabbed a movies dataset from Kaggle and augmented using some Python code and IMDB, and then retrained the index and voila!

Just added a more representative image and the initial verison is done! Ask away!

Testing in the app:

Question: "Who played Neo in the 'The Matrix'?" Answer: "Keanu Reeves played Neo in The Matrix" Sources: "data/movies.csv"

Question: "Who voiced the voice of Nemo in 'Finding Nemo'"?