Custom Layer Learnable Parameters Initialization.
2 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
David Ernesto Caro
am 8 Mär. 2020
Kommentiert: Jose Cortes-Briones
am 16 Jan. 2023
Hi!
I created a Custom Layer with Learnable Parameters (https://www.mathworks.com/help/deeplearning/ug/define-custom-deep-learning-layer.html)
However, to initialize these parameters I need to know the outputsize of the previous layer. I noticed that some pre-defined layers such as FullyConnected automaticly initialize its own weights using information from previous layers. Is there a way to automatize the weights initialization on custom layers at network initialization?
David
1 Kommentar
Jose Cortes-Briones
am 9 Sep. 2020
I asked a similar question a while ago and never got a response :/
Akzeptierte Antwort
Joss Knight
am 16 Jan. 2023
This is now supported via the optional method initialize: https://www-integ2.mathworks.com/help/deeplearning/ug/define-custom-deep-learning-intermediate-layers.html#mw_87362c7c-37d3-40c2-b5c8-c45ded304bcb
1 Kommentar
Weitere Antworten (1)
John Smith
am 4 Okt. 2021
Unfortunately, at least as of version 2021a, Matlab doesn't provide staright forward capabilities to automatically initialize learnable parameters of a custom layer similar to its built-in layers.
There's a workaround though using a shadow built-in network nested inside your custom layer (see here). Matlab will automatically initialize this layer from which learnable parameters you can pick up the values that you need.
It's a kludge, but it works (I've implemented such a scheme myself.
Siehe auch
Kategorien
Mehr zu Image Data Workflows finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!