Programs written while working through http://deeplearning.net/tutorial/deeplearning.pdf

Theano documentation is here: http://deeplearning.net/software/theano/#documentation

Original source code for book and data sources are here https://github.com/lisa-lab/DeepLearningTutorials.git

This is a place to store the code I write while working through "Deep Learning Tutorial"

My code is in Python 3, book and sample code is Python 2. Not sure which version of Theano he's using, I'm on 0.7/0.8, some minor changes were needed to get his code working running.

I'm also trying to keep each example net self contained instead of broken into separate files like book examples. This is so I can easily drop a network with only the parts I need into a new project. Once that's done it'll be easy to put the various classes back into separate files.

Completed:

Logistic Regression

Multilayer Perceptron

Convolutional ( LeNet )

Denoising Autoencoders

Stacked Denoising Autoencoders

Restricted Boltzmann Machine

Deep Belief Network

Hybrid Monte-Carlo Sampling

Elman-Recurrent (Python 2.x-3.x and replaced PERL script with slow but working Python)

Not yet here:

LSTM for Sentiment Analysis

RNN-RBM to model and sequence music