This repository contains all the project on Neural Networks, reports on experiences and interesting links.

• Simple Neural Network : A handcrafted 3 layer Network using biases, some evaluation and a report.

A 3 layer Neural Network. The input consists of 8 elements, the hidden layer uses 3 nodes and the output is 8 elements as well. Bias is added at the input and hidden layer, the sigmoid function is used for activation.

1. Change to the proper directory cd simple_neural_network
2. Install requirements with pip install -r requirements.txt
3. Run using python3 main.py

The implementation is mostly based on the Advanced Concepts in Machine Learning slides by Kurt Driessen

• http://outlace.com/Beginner-Tutorial-Backpropagation/
• https://triangleinequality.wordpress.com/2014/03/27/neural-networks-part-1/
• https://triangleinequality.wordpress.com/2014/03/31/neural-networks-part-2/
• http://www.bogotobogo.com/python/python_Neural_Networks_Backpropagation_for_XOR_using_one_hidden_layer.php
• https://medium.com/learning-new-stuff/how-to-learn-neural-networks-758b78f2736e#.tzmxfwvb6

• https://karpathy.github.io/neuralnets/
• https://medium.com/@ageitgey/machine-learning-is-fun-part-3-deep-learning-and-convolutional-neural-networks-f40359318721#.gdyx0872o