This is the code for "How to Do Math Easily - Intro to Deep Learning #4's challenge by Siraj Raval on youtube.

This is the code for this video on Youtube by Siraj Raval apart of the 'Intro to Deep Learning' Udacity nanodegree course. We build a 3 layer feedforward neural network trains on a set of binary number input data and predict the binary number output.

• Numpy

Run the demo.py script by running python demo.py in terminal.

• Used the earthquake magnitude prediction dataset.
• Added learning rate along with decay based on performance so that the learning curve can be controlled well.
• Split data into training batches which helped ease the computation to avoid dealing with very large matrices.
• Varied hyperparameters such as learning rate, number of hidden layers to measure and compare how the errors vary.
• Iterations per test varied based on the number of hidden units since larger networks take longer to train.

##Credits

Credits for the original code go to Andrew Trask