Abstract

We introduce a guide to help deep learning practitioners understand andmanipulate convolutional neural network architectures. The guide clarifies therelationship between various properties (input shape, kernel shape, zeropadding, strides and output shape) of convolutional, pooling and transposedconvolutional layers, as well as the relationship between convolutional andtransposed convolutional layers. Relationships are derived for various cases,and are illustrated in order to make them intuitive.

 

Quick Read (beta)