PyQt5 LLM Chat & Data Analysis GUI

For a detailed description of the project’s goals, architecture, and deliverables, see Project description.md.

Overview

This application is an interactive chat-based GUI for data analysis, powered by PyQt5 and a local LLM (Large Language Model) backend. It allows users to:

• Chat with an AI assistant
• Upload datasets (CSV or text files)
• Request statistical analysis and visualizations via natural language
• View results (including plots) directly in the chat interface
• Manage multiple chat rooms, each with its own context and file

Features

🗂️ Multi-Room Chat

• Create, switch, and delete chat rooms
• Each room maintains its own chat history and file context

📁 File Upload & Contextual Analysis

• Upload CSV or text files to a chat room
• The uploaded file becomes the context for that room
• The AI can answer questions and perform analysis based on the loaded file

💬 Natural Language Data Analysis

• Ask questions about the uploaded file in plain English
• Example: What is the mean of column A?
• Example: Show the time series for KO samples

📊 Interactive Plotting

• Request plots by describing them in chat
• Example: Plot histogram of column X
• Example: Plot the ratio of A to B
• Plots are generated using matplotlib and shown directly in the chat

📈 Ratio & Statistical Analysis

• Request ratio analysis between columns
• Example: Show the ratio of column1 to column2
• Get summary statistics, head, and description of the dataset

🧠 LLM-Powered Q&A

• If a file is loaded, the LLM uses it as context (RAG)
• If no file is loaded, the LLM answers general questions

🖥️ Modern PyQt5 GUI

• Responsive, resizable interface
• Status bar for feedback
• Clean separation of UI and logic for maintainability

Setup Instructions

1. Clone the repository

2. Install dependencies (in your virtual environment):

   pip install -r requirements.txt

3. Set up your local LLM backend (e.g., Ollama)

   • See Set Up Ollama.md for platform-specific installation and setup instructions.

4. Run the application:

   python main.py

Usage

1. Start the app: python main.py
2. Create or select a chat room
3. Click Load File to upload a CSV or text file
4. Type your analysis request (e.g., plot histogram of column X, show the ratio of A to B)
5. View results and plots in the chat

Example Requests

• Plot histogram of column temperature
• Show the ratio of pressure to volume
• What are the main differences between OK and KO samples?
• Describe the dataset

Project Structure

main.py
logic/
  llm_handler.py
  ...
gui/
  chat_window.py
  room_list_panel.py
  chat_display_panel.py
  input_panel.py
  ...
dataset/
resources/
tests/

Requirements

• Python 3.8+
• PyQt5
• pandas, matplotlib
• langchain, ollama, faiss, etc. (see requirements.txt)

License

MIT