Calculate the output Size of a CNN (Convolution Neural Network) layer

default-2023

input_width

input_height

Input Channels (D):

kernel_width

Filter Count (K):

vertical_dilation

horziontal_dilation

horizontal_stride

horizontal_padding

filter

input_channels

kernel_height

vertical_stride

vertical_padding

Kernel Size (F): Size of the convolving kernel (Height, Width)Filter Count (K): The number of filters in the convolving layerStride (S): Step size of the filter. Default: 1 (Vertical, Horizontal)Padding (P): Padding added to all four sides of the input. Default: 0 (Vertical, Horizontal)Dilation (d): &nbsp;Spacing between kernel elements. Default: 1 (Vertical, Horizontal) Equations UsedHout = {{"\[\\frac{H_{in} + 2 * P[0] - d[0] * (F[0] - 1) - 1}{S[0]}\] + 1" | katex}}Wout = {{"\[\\frac{W_{in} + 2 * P[1] - d[1] * (F[1] - 1) - 1}{S[1]}\] + 1" | katex}}Dout = K

Input Size

Convolution

How to calculate output size of a convolution layer

2D CNN layer output size calculator

Similar apps: