Keras is a high-level API capable of running on top of TensorFlow, CNTK, Theano, or MXNet (or as tf.contrib within TensorFlow). Since its initial release in March 2015, it has gained favor for its ease of use and syntactic simplicity, facilitating fast development. It’s supported by Google.
PyTorch, released in October 2016, is a lower-level API focused on direct work with array expressions. It has gained immense interest in the last year, becoming a preferred solution for academic research, and applications of deep learning requiring optimizing custom expressions. It’s supported by Facebook.
These are a set of links I used when learning about machine learning and neural networks: