DistETA: Distributional ETA Analysis & LLM Reporting

This project provides a robust, two-part pipeline for analyzing time-based operational data and generating insightful, human-readable reports using Large Language Models (LLMs). It moves beyond simple averages to understand the full distribution of outcomes, identifying distinct, recurring operational patterns.

This software is provided for demonstration and evaluation purposes only. It is not intended for commercial use. See the LICENSE file for more details.

For a detailed explanation of the project's workflow, architecture, and configuration, please see the Detailed Workflow Documentation.

Getting Started

Prerequisites

• Python 3.10+
• Docker Desktop (recommended)
• An API key for Google Gemini (if using the Gemini provider).
• Ollama installed and running with a multimodal model (if using the Ollama provider).

Installation & Usage (Docker)

1. Clone the repository
2. Create a .env file in the project root and add your Google API key:

   GOOGLE_API_KEY="your_api_key_here"
   

3. Run the full pipeline (analysis and reporting):

   docker-compose --profile analyze_report up --build

4. View the report at http://localhost:5001.

Local Installation

1. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate

2. Install dependencies:

   pip install -r requirements.txt

3. Run the analysis:

   python -m src.disteta_batch.main

4. Generate the report:

   python -m src.report_generator.main

Project Structure

• config/: YAML configuration files for the analysis and reporting.
• data/: Input data files.
• docs/: Detailed project documentation.
• output/: Generated reports and analysis results.
• src/: Main source code.
• tests/: Pytest tests.