KaggleGalaxyZoo | My submission to the Kaggle competition Galaxy Zoo

KaggleGalaxyZoo is a Python library. KaggleGalaxyZoo has no bugs, it has no vulnerabilities and it has low support. However KaggleGalaxyZoo build file is not available. You can download it from GitHub.

My submission to the Kaggle competition "Galaxy Zoo". The goal of the competition was to predict the probability distributions of geometric properties of galaxies from their images. My best submission produced a score of ~ 0.107 and landed me at 65th position at the end of the competition, under the team name Milchstraße. Was ok with my performance given that this was my first time working with images and also that I did not know about ConvNets back then. The data for the competition can be downloaded from The train.py code will have to be run first. This will extract various geometric, segment related and colour based features from the images using computer vision techniques. The best subset from these features are used to train a single layer neural network, one for each output class to be predicted. These models are pickled to the data folder. Next predict_test_images.py is run which extracts the necessary features from the test images and re-loads the pickled models and makes the final predictions using these.