How to calculate the output in Convoluted Neural Networks?

machine_learning
deep_learning

#1

I read Convolutional Neural Networks in web :http://cs231n.github.io/convolutional-networks/ But i don’t understand how to caculate output as image below . Any one can help me ?


#2

You have to multiply each data point with the Filter W1 matrix.
For the case which is marked:-
Input Volume 1 has selected matrix which will be multipled with filter W1
Now if you do multiplication element wise like x[0,0] * W1[0,0] = 0 * (-1) = 0
x[0,1] * W1[0,1] = 0 * 0 = 0
x[0,2] * W1[0,2] = 0 * 1 = 0
x[1,0] * W1[1,0] = 2 * 0 = 0
x[1,1] * W1[1,1] = 2 * -1 = -2
x[1,2] * W1[1,2] = 0 * 1 = 0
x[2,0] * W1[2,0] = 2 * -1 = -2
x[2,1] * W1[2,1] = 2 * -1 = -2
x[2,2] * W1[2,2] = 0 * 1 = 0

Now if you sum this up you will get :- -6

Now repeat it for x[:,:,1] and w1[:,:,1] where you will get -4

and for x[:,:,2] and w1[:,:,2] you will get +3

Now if you add this up you will get -6+(-4)+3 = -7

Hope this helps


#3

thank you so much