Take a Course!
This link is a good start. There's lots of them.
Read a Book!
I can highly recommend Deep Learning: A Practioner's Approach. From a preview of this, it's extremely well-written and easy to follow — and includes lots of code examples. It is by the brilliant minds of the Deep Learning 4J people.
They wrote that.
The Big Ones: MXNet, TensorFlow, and Keras
Tutorials of Keras: This could be the Rosetta Stone, or at least the Apache Beam, of DL.
Deep Learning Meetup: Good comparison of a few frameworks.
Autopilot: TensorFlow: This is some great code to try and learn from.
Agile Data Science: This is the way to get it done.
Step 0: Learn Python. Learn Python. Use Python.
Early Preview of PyTorch
Time for me to OSX it!
pip install https://s3.amazonaws.com/pytorch/whl/torch-0.1.9.post2-cp27-none-macosx_10_7_x86_64.whl pip install torchvision
Every good framework needs examples and tutorials, especially when it is something uber complex like Machine Learning or Deep Learning. Pytorch is new but already has a good introductory set of examples and tutorials to learn from and expand to.
Yet Another Framework and Variation, Chainer v2
Install easily in Python with PIP:
pip install chainerrl