This repo contains a local version of the practical notebooks created by Deeplearning.ai as part of their training, which is available at:
https://learn.deeplearning.ai/langchain-chat-with-your-data
The code was partially modified to run completely locally in a laptop/desktop device using CPU and the Open source Ollama Mistal LLM, as Python code instead of Jupyter notebooks.
Original documents of the course were not Downloaded, and filenames are hardcoded on the code.
Getting a valid API from smith.langchain.com is recommended to use QAdebug.py
No OpenAPI is required to run these examples.
No OpenAPI embeddings are used is required to run these examples. Instead, hkunlp/instructor-base
from NLP Group of The University of Hong Kong are used.
- Deeplearning.ai and Harrison Chase
- NLP Group of The University of Hong Kong
- Ollama implementation https://ollama.ai/
- Hugging Face https://huggingface.co/
This is a learning exercise and code is provided as is.