How can I continue training a fitrnet network?

7 Ansichten (letzte 30 Tage)
AC Fain
AC Fain am 14 Okt. 2021
Beantwortet: Sahil Jain am 20 Okt. 2021
I am training a regression neural network using Matlab's fitrnet function. I want to know whether there is a way to continue training a fitrnet network after it has completed? If I train a neural network using Matlab's trainNetwork function then to continue training all I need to do is use the previously trained network's layers as the layer argument (ex. trainNetwork(inputData, outputData, old_net.Layers, opts) ). Is there a way to do something similar with fitrnet? I find fitrnet works a lot better for my application that trainNetwork

Akzeptierte Antwort

Sahil Jain
Sahil Jain am 20 Okt. 2021
As of R2021b, the "fitrnet" function does not support loading weights and biases before training. As a workaround, you can modify the training options for "trainNetwork" to achieve results similar to "fitrnet".

Weitere Antworten (0)

Community Treasure Hunt

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

Start Hunting!

Translated by