How to know the number of filter

Dear all, I want to know the number of convulation that used in deep learning. Because there is no number show. The padding and stride have. But filter not have. Any one can explained

.

Answer 1

[Image Analyst]

These filters can be done by imfilter or conv2. If the stride, which I assume to be the number of pixels the window jumps by to get to the next location, is greater than 1, you can use blockproc. See attached demo.

The "number of convulation that used in deep learning" depends on the network. Examine the network structure to learn that.

Not sure what you mean by padding, but you can pad arrays with padarray.

[mohd akmal masud]

what I mean is, when I try to design my network trained, when selected convulation, the number filter inside does not show. as picture in red box, just have number of filter anf size of filter.

but number filter in it not have, like

1 1 1

0 0 0

1 0 1

[Image Analyst]

Of course not. The weights inside the convolution window are what the Deep Learning figures out. They change as training goes on and will bascially look like random numbers when it's finished. That's how training a network works.

Really though, you do not need to know what they are.

[mohd akmal masud]

oh i see...tq image analyst