I recommend using anaconda3

• python libraries
  • keras (>1.0)
  • theano
  • pydot_ng
  • matplotllib
  • daft (only tutorial )
  • gensim (only save and load word embeddings)
• software
  • graphviz

• Image
  • MLP_MNIST
  • CNN_MNIST
  • CNN_CIFAR10
  • autoencoder
  • variational auto encoder for CIFAR10
  • Gumbel-softmax with variational auto encoder for MNIST
  • residual CIFAR10
• Text
  • RNN_binary_classification
  • encoder-decoder
  • NNLM
  • CBoW : not hieraltical softmax and negative sampling
  • Skip-Gram : not hieraltical softmax and negative sampling
  • Skip-Gram&NG : using negative sampling
  • fast text : not hieraltical softmax and bi-gram features
  • SCRNLM : Structurally Constrained Recurrent Nets Language Model
• Tutorial
  • keras_tutorial : for Machine Learning seminar at Tsukuba (note: 7MB notebook and ja only)