This repository contains a collection of Jupyter notebooks that provide an analysis and comparison of three prominent vector databases: Pinecone, FAISS and pgvector. These notebooks summarize my first experience and evaluation of these databases as part of a pet project named "DRY" (Do Not Repeat Yourself).
The DRY project focuses on enhancing group chatbots by incorporating automatic background semantic search capabilities. By leveraging the power of vector databases, DRY aims to intelligently identify and respond to repetitive questions by retrieving similar messages from previous conversations.
To run these 3 notebooks, you may try accessing them through Google Colab:
or you can import the .ipynb files from this repository into a new Google Colab or other Jupyter notebook. Subsequently, please refer to the instructions provided within the notebooks and/or the accompanying Youtube video for guidance.