Why do we pass input X to a layer like BatchNormalization like this : X = BatchNormalization(axis = 3)(X)

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

Why do we pass X from the previous layer into BatchNormalization layer like this X = BatchNormalization(axis = 3)(X). can’t we directly pass X as a argument? Also does X = BatchNormalization(axis = 3) return a function to which we pass X??

2

Maybe take a look at this thread and let us know what you discover ?

1 Like