Define OpenAI and Langchain API keys in a file named .env.
The required parameters are shown in the example file .env.example.
This repository contains example notebooks demonstrating how to use Langchain with OpenAI.
| Notebook | Description |
|---|---|
| translate.ipynb | Notebook demonstrating translation using user prompts with Langchain and OpenAI. |
To set up PostgreSQL with AI capabilities, follow these steps:
-
Install the required extensions by running the SQL commands provided in
database/vector_and_ai_extensions_install.sql. These extensions include:- pgvector: Enables vector similarity search in PostgreSQL, essential for AI-driven applications.
- ai and vectorscale: Enhance PostgreSQL's ability to handle AI workloads efficiently.
-
Refer to this tutorial on YouTube for a great guide of working with these extensions.
By integrating pgvector and other tools, this project enables efficient storage and querying of vector embeddings, making it suitable for AI and machine learning workflows.
- Clone the repository.
- Set up your
.envfile based on the provided.env.example. - Open the notebook in Jupyter or VSCode and follow the instructions inside.
https://github.com/Bessagg/Langchain_examples
This project uses data from EHR-CON: Consistency of Notes (version 1.0.0), a small example database provided by PhysioNet. This dataset was selected due to its manageable size, making it suitable for development and testing purposes before scaling to larger datasets like MIMIC-IV. Dataset link: https://www.physionet.org/content/ehrcon-consistency-of-notes/1.0.0/
This project utilizes data from MIMIC-IV-Note, a publicly available dataset of deidentified clinical notes: Johnson, A., Pollard, T., Horng, S., Celi, L. A., & Mark, R. (2023). MIMIC-IV-Note: Deidentified free-text clinical notes (version 2.2). PhysioNet. https://doi.org/10.13026/1n74-ne17