GitHub - jiega/MNIST-with-Feed-Forward-Neural-Network-from-scratch: implemented a Feed-Forward Neural Network from scratch to classify the MNIST data set. Made improvements by using mini-batch gradient descent, Kaiman-initialization to address the exploding and vanishing gradient problem, and ReLU as activation function.

jiega / MNIST-with-Feed-Forward-Neural-Network-from-scratch Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

implemented a Feed-Forward Neural Network from scratch to classify the MNIST data set. Made improvements by using mini-batch gradient descent, Kaiman-initialization to address the exploding and vanishing gradient problem, and ReLU as activation function.