BDH (Dragon Hatchling) – Architecture and Code

This project focuses on image compression using a convolutional autoencoder. The goal is to compress satellite images from the EuroSAT dataset into a compact representation and reconstruct them wit…

As discussions around asteroid risks, like 2024 YRA, continue to rise, I took it upon myself to explore how machine learning could help assess the danger of near-Earth objects NEOs. By using a ML, …

ML toolkit for analyzing the effective use of weights in neural networks. It provides quantitative metrics (entropy, coverage, FLOPs) and visualizations to identify redundant parameters and guide m…