We will cover series of generative ai tasks which include but not limited to:
- Retrieval-Augmented Generation (RAG)
- LLM Page Summarization
- Retrieval-based Chatbots
- so on...
RAG is a process of fetching up to date or context specific data from an external database and making it available to to an LLM when asking it to generate a response.
To be able to do this, we need an open source language model, a vector database and a composer. Fortunately, there are freely available open source python libraries to create this solution. For simplicity, we will use the following:
- Pre-trained T5 model from Huggingface as LLM
- ChromaDB as vector database
- Langchain as application tools.
Click here for more.
Instead of using a pre-trained T5 model, we will use gpt4all models.
Click here for more.
To summarize a page, we will use a GPT4All as LLM.
For more, click here
Simple but functional Chatbot.
Click here for more.
I'm a full stack AI developer...
