Is it possible to use non-image data as inputs for the Deep Network Designer App?

2 visualizaciones (últimos 30 días)
I have a 3xn vector of inputs that I would like to use with the Deep Network Designer app. However, it seems that the input layers are generally geared towards image classification. Is there a way to use the numerical data in matrix format as inputs?
Thank you.

Respuestas (1)

Ive J
Ive J el 26 de Ag. de 2021
Technically yes you can, I assume something like this would work:
dlData = reshape(data', [1, 1, size(data, 2), size(data, 1)]);
then your layers would be something like:
layers = [imageInputLayer([1 1 size(data, 2)]);
% other layers
];
  1 comentario
Michael McGeehan
Michael McGeehan el 30 de Ag. de 2021
Thanks for your response. The data options for the input layer require me to specify a folder with images. Is there any way around this?

Iniciar sesión para comentar.

Categorías

Más información sobre Image Data Workflows en Help Center y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by