How to know the correct amount of filters to apply in each convolutional layer?

In the examples 3 convolutional layers are added with 16,32 and 64 filters, is there a rule to set the number of filters?
The number of filters depends on the size of the images?
The minimum number of filters is 16?

Hi @Luis_Anibal_Vasquez,
the dimension of the image can influence the number of filters, but mostly is the number of features that you want to catch. The higher this number, the higher the filters. Of course, when the image is bigger, you may need to find more features. My suggestion is to try different configurations and see what works well.


Hi @maurizioscibilia, thank you so much for the information, I’ll keep in mind your suggestion!!