Implementation of the SDA using theano library.

To run the SDA for the horse racing dataset please follow the guide from below. You can also edit the test_sda.py if you want to tune the parameters.

1. ./setup.py
2. ./test_sda.py

• data/training.csv - the horse racing dataset for training
• data/testing.csv - the horse racing dataset for testing
• setupy.py - a script preprocessing the data/testing and data/training into a
• format to be used for machine learning
• test_logistic_regression.py - a script for testing logistic regression
• test_denoising_autoencoder.py - a script for testing denoising autoencoder
• test_sda.py - a script for testing stacked denoising autoencoder
• deeplearn/classifers.py - classifiers implementation
• deeplearn/metrics.py - confusion matrix
• deeplear/datasets.py - utils for generating datasets