Introduction to deep learning based on the Keras framework. These tutorials are direct ports of Nathan Lintz's TensorFlow tutorials.

The Keras tutorials are best run within Docker. First, build the Docker container:

$ docker build -t keras-notebook .

Run the non-GPU container using:

docker run -it -p 8888:8888 keras-notebook

The Docker image extends the official TensorFlow Docker image. Currently, the image does not support GPU containers and is not on Docker Hub (will update soon).

• Matrix Multiplication
• Linear Regression
• Logistic Regression
• Feedforward Neural Network (Multilayer Perceptron)
• Deep Feedforward Neural Network (Multilayer Perceptron with 2 Hidden Layers and Dropout)
• Convolutional Neural Network
• Autoencoder
• Recurrent Neural Network (LSTM)
• Word2vec
• Transfer Learning
• Save and Load a Neural Network