MLP network in Keras to classify stellar objects in 3 classes given a set of numeric features.
Project build in a Jupyter Notebook. Main file is Rede.ipynb. Dependencys include Keras, tensorflow and sklearn. All the data used to train the network is in the CSV file. The Network build can classify stellar objects in Galaxy, Quasar and Star. The input used is a set of 17 features as shown in the CSV. In the notebook there is the data preprocessing and the whole Network trainment which is very fast, take very few seconds in a notebook NVidia GPU.