Predicting the input for the next layer

I noticed that in the assignment we had a sketch displayed of the neural network. I was just a little bit confused as to how do we look at a dataset and figure out the number of inputs each layer will take?

Hi @ojgupta4321

In this notebook assignment or next assignment ,they told that what each layer should
*layer1: The shape of W1 is (400, 25) and the shape of b1 is (25,)
*layer2: The shape of W2 is (25, 15) and the shape of b2 is: (15,)
*layer3: The shape of W3 is (15, 1) and the shape of b3 is: (1,)

but in the real live project you determine what each layer should base on for example in the input layer neurons should be equal number of input feature(input feature should be flatten) any you decide what is the number of neurons in next layers or number of layers base on how do you want NN be complex and more effective but when you decide to increase number of layers and number of neurons in next layers it cost much time

Please feel free to ask any questions,