3 exampels demonstrating what neural networks are capable of and how to implement them with Python using libaries such as tensorflow and scikit-learn
Instead of using machine learning to train models like in Learning, this implementation uses the tensorflow suquencial neural network to train a network that identifies counterfeit banknotes from published ones
Demonstrate how neural networks can be applied to image processing, implements the image convolution which can be added to a convolutional neural network. This specific implementation uses the famous [-1, -1, -1, -1, 8, -1, -1, -1, -1] filter which results in a new image with only the shape of objects in the original image
Using the famout MNIST dataset, train a neural network to identify handwritings of digits 0-9. Comes with a Pygame that allows user to handwrite digits and the neural network will identify the number user wrote