We introduce a guide to help deep learning practitioners understand and
manipulate convolutional neural network architectures. The guide clarifies the
relationship between various properties (input shape, kernel shape, zero
padding, strides and output shape) of convolutional, pooling and transposed
convolutional layers, as well as the relationship between convolutional and
transposed convolutional layers. Relationships are derived for various cases,
and are illustrated in order to make them intuitive.(more)
Please log in to take part in the discussion (add own reviews or comments).
Cite this publication
More citation styles
- please select -
%0 Generic
%1 dumoulin2016guide
%A Dumoulin, Vincent
%A Visin, Francesco
%D 2016
%K
%T A guide to convolution arithmetic for deep learning
%U http://arxiv.org/abs/1603.07285
%X We introduce a guide to help deep learning practitioners understand and
manipulate convolutional neural network architectures. The guide clarifies the
relationship between various properties (input shape, kernel shape, zero
padding, strides and output shape) of convolutional, pooling and transposed
convolutional layers, as well as the relationship between convolutional and
transposed convolutional layers. Relationships are derived for various cases,
and are illustrated in order to make them intuitive.
@misc{dumoulin2016guide,
abstract = {We introduce a guide to help deep learning practitioners understand and
manipulate convolutional neural network architectures. The guide clarifies the
relationship between various properties (input shape, kernel shape, zero
padding, strides and output shape) of convolutional, pooling and transposed
convolutional layers, as well as the relationship between convolutional and
transposed convolutional layers. Relationships are derived for various cases,
and are illustrated in order to make them intuitive.},
added-at = {2021-01-20T13:07:25.000+0100},
author = {Dumoulin, Vincent and Visin, Francesco},
biburl = {https://www.bibsonomy.org/bibtex/24282c5ae16fe6012c818a9dc7c442a0d/philipphaas},
interhash = {47f037084e2155cec24f09bf4897dcba},
intrahash = {4282c5ae16fe6012c818a9dc7c442a0d},
keywords = {},
note = {cite arxiv:1603.07285},
timestamp = {2021-01-20T13:07:25.000+0100},
title = {A guide to convolution arithmetic for deep learning},
url = {http://arxiv.org/abs/1603.07285},
year = 2016
}