ivotomasovich / ollama-url-driven-chatbot Goto Github PK

Document Query Chatbot with Ollama Mistral

Overview This project showcases a chatbot built using the Ollama Mistral model. The chatbot can process documents from specified URLs and answer questions based on the content of those documents. The system uses Retrieval-Augmented Generation (RAG) to enhance the accuracy and relevance of the responses. The project is implemented in two Python files: app.py and ui.py.

Files app.py This script demonstrates the core functionality of the chatbot:

Data Loading: Loads documents from specified URLs. Text Splitting: Splits documents into manageable chunks. Embedding Generation: Converts document chunks into embeddings and stores them in a vector store. Before and After RAG: Shows the chatbot's response without context (Before RAG) and with context (After RAG).

ui.py This script provides a user interface using Gradio:

User Input Processing: Takes URLs and a question as input. Data Loading and Processing: Loads documents, splits them, and stores embeddings similar to app.py. RAG Implementation: Answers the question based on the context extracted from the documents. Gradio Interface: Provides a simple web-based UI for users to interact with the chatbot.

How to Run Install Dependencies: pip install langchain_community gradio

Run the App: For the core functionality: python app.py

For the user interface: python ui.py

What I Learned Language Models: Gained hands-on experience with the Ollama Mistral model for natural language processing tasks. Data Processing: Learned to handle and process large textual data by splitting documents into chunks and generating embeddings. Retrieval-Augmented Generation (RAG): Implemented RAG to improve the relevance of the chatbot's responses, demonstrating the importance of context in NLP applications. Web-based Interfaces: Utilized Gradio to create a user-friendly interface, making it easier for users to interact with the chatbot.