Project Overview
Evaluated the effectiveness of Generative Adversarial Networks (GANs), Conditional GANs (CGANs), and Contrastive GANs (ContraGANs) for generating synthetic data to predict the B-V color index in astronomical datasets. Assessed models based on predictive accuracy, clustering quality, and statistical alignment. ContraGAN outperformed other models with the lowest MSE of 0.1294, highest R² of 0.7258, and superior clustering scores. Highlighted the benefits of contrastive learning and conditional generation for synthetic data applications in scientific and industrial fields.
Key Features
Data Source: Utilizes photometric data from over 27 million galaxies collected by SDSS-DR7.
Algorithms Implemented
Linear Regression
Long Short-Term Memory (LSTM) Networks
Support Vector Regression (SVR)
Random Forest Regressor
Decision Tree Regressor
Gradient Boosting Regressor
Classical Deep Learning Models
Performance Metrics: Evaluates model performance using Mean Absolute Error (MAE) and other relevant metrics. Comparative Analysis: Provides a comprehensive comparison of the performance of different algorithms in estimating SFRs from photometric data.
Contributing
Contributions are welcome! Please submit a pull request or open an issue if you have any suggestions or improvements.
Contact
For any questions or inquiries, please reach out to [satvikraghav007@gmail.com].