Deep learning is a set of neural networks algorithms inspired by the structure and function of the brain. This session provides a hands-on introduction to deep learning using TensorFlow and Keras in Python and R programming languages. TensorFlow is Google’s deep learning engine. Keras is a high-level neural networks API. Because of its easy to use, it has been adopted as an interface for TensorFlow, CNTK, and Theano. This presentation will begin with FAQ and an introduction to some basic concepts of deep learning, then jump start with an example of a small neural network training to classify images. The example demonstrates: - how to prepare the data; - how to create a deep learning model; - how to specify layers with ReLU or softmax activation functions; - how to train the model; - how to evaluate the accuracy; - how to predict with the model. The example is written in Python and R identically.