Neural Networks (E02: predictions – python)
In this episode we look at how neural networks can be represented with matrices, and create a simple feedforward network in python.
Note: at 2:31 the second bias vector should only have two rows, not three.
I owe a lot to this excellent online book on neural networks:
Learn more about weight initialization:
(note: numpy.random.randn and numpy.random.standard_normal are functionally equivalent, the latter just takes a tuple for the shape parameter).