Overview
Introduction
GalaxyHackers is a state-of-the-art framework designed for astronomical data analysis, with a primary focus on galaxy cluster detection. The project provides tools for building, training, and evaluating machine learning models specifically tailored to the unique challenges of astrophysical datasets.
Key features include:
Model Variety: Supports a wide range of architectures, including ResNet, EfficientNet, DenseNet, AlexNet/VGG, SpinalNet, CNN-MLP, and a baseline model. Both popular and custom implementations are available for flexible experimentation.
Automatic Segmentation Visualization: Easily generate visual outputs for cluster detection results.
Comet ML Integration: Monitor and analyze training experiments seamlessly.
Extensibility: Easily add new datasets, models, or workflows with minimal setup.

Architecture Overview
The framework follows a modular architecture, making it easy to integrate new components while keeping the codebase maintainable.
Modules include:
Data: - Handles dataset preparation and augmentation.
Models: - Predefined and custom machine learning models.
Train: - Utilities for model training and validation.
Metrics: - Compute evaluation metrics such as accuracy, precision, recall, and segmentation.