How can we use Nadam optimizer in place of sgdm in training deep learning networks

4 vues (au cours des 30 derniers jours)
Training_Options = trainingOptions('sgdm', ...
'MiniBatchSize', 32, ...
'MaxEpochs', 50, ...
"InitialLearnRate", 1e-5, ...
'Shuffle', 'every-epoch', ...
'ValidationData', Resized_Validation_Data, ...
'ValidationFrequency', 40, ...
"ExecutionEnvironment","gpu",...
'Plots','training-progress', ...
'Verbose',false);

Réponses (1)

Joss Knight
Joss Knight le 4 Avr 2023
You cannot do this using trainNetwork. You need to use a dlnetwork with a custom training loop so you can author your own update rule. Perhaps adam will work for you instead.

Catégories

En savoir plus sur Image Data Workflows dans Help Center et File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by