Anyone Can Learn to Code an LSTM-RNN (Part 1: RNN)

If you want to know more about RNNs in general, I can't recommend watching the videos/reading the notes from this course enough:

http://cs224d.stanford.edu/syllabus.html

If you want something more basic to get your head around NNs, I recommend Denny Britz's "Neural Networks from Scratch":

http://www.wildml.com/2015/09/implementing-a-neural-network-...

I created a gist with a heavily commented version of his code:

https://gist.github.com/sthware/c47824c116e6a61a56d9

For a more thorough intro, I can't recommend anything better than this http://arxiv.org/abs/1506.00019

This is a wonderful post. For anyone who'd like to see a nice intro to decision trees insofar as how they apply to machine learning, this is a nice post as well: http://www.r2d3.us/visual-intro-to-machine-learning-part-1/

Another fantastic resource, from a programmer's perspective. Very little math. http://karpathy.github.io/neuralnets/

In theory yes, but these NN libraries are very difficult to set up. I remember it was a huge pain to get numpy working with lots of frustrations and learning technical things. Torch and theano are not supported on Windows at all.

I've been looking at brain simulator recently. It has it's own installer and a decent GUI: http://www.goodai.com/#!brain-simulator/c81c

Some more information on LSTMs http://colah.github.io/posts/2015-08-Understanding-LSTMs/

Terrific post! Its great to see things implemented in short, 1-file programs. Tends to demystify things a lot.

What did you use to make the diagrams + gifs?

I can't if I can't access that link.