MNIST Handwritten Digit Classifier – beginner neural network project

Re:MNIST Handwritten Digit Classifier – beginner neural netw... by iopq on Sunday July 31, 2016 @09:50AM (#12196958)

I thought Neural Networks nowadays use ReLU instead of Sigmoid? Especially in the context of deep learning

[ Parent | on HN ]

Re:MNIST Handwritten Digit Classifier – beginner neural netw... by asib on Sunday July 31, 2016 @10:30AM (#12197100)

Looks like this implementation is based (in part) on the Stanford ML course, which teaches nnets using sigmoid activation.

Given that it's intended to introduce to beginners how nnets work, the choice of activation is an aside anyway - the real meat is back/forwardprop.

[ Parent | on HN ]

Re:MNIST Handwritten Digit Classifier – beginner neural netw... by ironislands on Sunday July 31, 2016 @10:56AM (#12197219)

Absolutely, though I'll put in softmax, tanh and ReLU activation functions in near time. It isn't that difficult. Putting in more docstrings is also one of my todo.

[ Parent | on HN ]