Ask questions to your markdown documents:
- Upload your markdown documents
- They are converted to embeddings using Chromadb
- You ask a question
- Your question is converted to embedding and the app gets the more similar documents from the vector database
- The app asks OpenAI API to answer the original question, providing the retrieved documents.
- You have your answer!
![Screen Shot 2023-05-27 at 19 11 27](https://private-user-images.githubusercontent.com/1221345/241389402-d0e6ae86-0364-4046-ba09-f9f71afffdae.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.a6XFvbq104wSQozqum5R9037qQ-rMmS_-fTYQyLmQdg)
Create a .env file with your OpenAI Api key, and the url of your python server, and put it inside webapp
folder.
Your file should look like this:
SECRET_OPENAI_API_KEY="sk-XXXX"
PUBLIC_PROCESS_FILES_SERVER="http://127.0.0.1:5000"
This project is explained in detail here:https://www.youtube.com/watch?v=rtX8oqolN8s (Spanish)