CNN for Galaxy Classification + Webapp

This repository provides an implementation of a Convolutional Neural Network (CNN) for classifying galaxy morphologies using the Galaxy10 dataset. The project utilizes TensorFlow for building and training the model and follows standard preprocessing steps to ensure efficient data handling and model performance.

The accompanying webapp based on this model can be found here.

TL;DR

This repository contains a Jupyter notebook (Final_CNNN(1).ipynb) that trains and evaluates a CNN for image classification. The notebook includes data loading, model definition, training loop, evaluation (metrics & confusion matrix), and sample inference cells. Detected frameworks/tools used: TensorFlow/Keras, model checkpoints.

Libraries Uses

jupyterlab
numpy
pandas
matplotlib
scikit-learn
Pillow
opencv-python
tensorflow/keras        
seaborn