MNIST

Different ML models for MNIST dataset predictions

1. NearestNeighbors.py uses K-Nearest Neighbors lazy-learning method
2. DecisionTrees_RandomForest.py finds first the best DecisionTree configuration and then trains a RandomForest with such characteristics
3. NeuralNetworks.py builds a Deep Neural Network model either with a Feed Forward Neural Network or a Convolutional Neural Network
4. AutoEncoders.py doesn't aim at predicting labels like the other models; it tries instead to extrapolate the defining features of the inputs and to recreate its representation