Question

Suppose you have a 10x10x3 colour image input and you want to stack two convolutional layers with kernel size 3x3 with 10 and 20 filters respectively. How many parameters do you have to train for these two layers? Don't forget bias terms!

Answer 1

We have,

10 * 10 * 3 color image input and to convolutional layers with kernel size 3 * 3 with 10 and 20 filters.

Number of parameters required:

= (3 * 3 * 3 + 1) * 10 + (3 * 3 * 3 + 1) * 20

= (28) * 10 + (28) * 20

= 280 + 560

= 840.

Therefore, 840 parameters you have to train for these two layers.

Hope it helps!

Answer 2

Answer:

2100

Explanation:

#parameters = (F.F.D).K + K

:F.F.D : Number of parameters for each filter (analogous to volume of the cuboid)

:(F.F.D).K : Volume of each filter multiplied by the number of filters

:+K: adding K parameters for the bias term

So,

(3*3*3+1)*10 + (3*3*10+1)*20

=280+1820

=2100