Spectrosense

🔄 Migration Guide
If upgrading from v0.2.x:
1. Update your .env file with new AI configuration options
2. Choose your preferred model type
3. Install additional dependencies if using Llama:
bash
pip install vllm


🔍 Notes
- Llama model requires significant GPU resources
- vLLM server must be running for Llama integration
- Claude integration requires valid API key

🔍 What's New
Introducing direct spectrogram image analysis! SpectroSense now supports analyzing existing spectrogram images without needing the original signal data. Just point it at your spectrogram images and let our AI do the work.

Key Features
- Direct analysis of `.png`, `.jpg`, `.jpeg`, `.tiff`, and `.bmp` spectrogram images
- Batch processing capability for multiple images
- Automatic image preprocessing and validation
- Detailed JSON reports with signal classification
- Support for high-resolution spectrograms up to 1920x1080
- Comprehensive error handling and validation

Technical Improvements
- New `SpectrogramImageAnalyzer` class for dedicated image analysis
- Enhanced preprocessing pipeline for optimal AI analysis
- Batch processing support with detailed reporting
- Improved error handling and file validation
- Maintains original file structure while adding analysis outputs

Usage Example
python
from spectrosense import SpectrogramImageAnalyzer

analyzer = SpectrogramImageAnalyzer(api_key="your_api_key")
result = analyzer.analyze_spectrogram_image("spectrogram.png")


📦 Installation
bash

🔄 Upgrading
bash
pip install --upgrade spectrosense


Would you like me to expand on any part of the release notes or add additional sections?