How to train neural network using newff with a custom loss function?

15 visualizaciones (últimos 30 días)
Rah Yoonhyuk
Rah Yoonhyuk el 4 de Feb. de 2020
Respondida: Hiro Yoshino el 4 de Feb. de 2020
I want to train a neural network (f) with input A and output B. I want to train it with the loss function between g(f(A)) and C where g is a pre-trained neural network. C is another data set. How can I use a custom loss function? Are there any other functions other than newff? Below is my code (very simple)
net = newff(minmax(A),[10,51],{'logsig','purelin','trainlm'});
net.trainparam.goal = 0;
net.trainparam.lr = 0.01;
net = train(net, A, B);

Respuestas (1)

Hiro Yoshino
Hiro Yoshino el 4 de Feb. de 2020
Which version are you using?
Looks like there is no such a function.

Categorías

Más información sobre Deep Learning Toolbox en Help Center y File Exchange.

Productos


Versión

R2019a

Community Treasure Hunt

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

Start Hunting!

Translated by