This folder is intended as a playground for experimenting with various neural network models and libraries. It contains implementations of

• mnist_mlp: A simple multilayer perceptron for MNIST implemented with keras

• mnist_cnn: A simple convolutional neural network for MNIST implemented with keras

• usps_cnn: A simple convolutional neural network for USPS dataset implemented with keras

• variational_autoencoder: Two implementations (one in pure Theano, one in lasagne) of the model proposed in

  Kingma, D. P., & Welling, M. (2013). Auto-Encoding Variational Bayes. arXiv:1312.6114