Can trainNetwork handle Multiple Mini-Batch Datastores?

3 views (last 30 days)
Is it possible to use all of following Mini-batch Datastore functions to Train my neural network: augmentedImageDatastore, pixelLabelImageDatastore, denoisingImageDataStore

Answers (1)

Sayan Saha
Sayan Saha on 7 May 2018
According to the documentation for trainNetwork each of the datastores can be used for training with "trainNetwork". You can refer to the documentation for the individual datastores for examples showing how to use "trainNetwork" with these datastores.
If on the other hand, your use-case requires a combination of these datastores, then please refer to how to develop custom datastores to build a custom one that had the functionalities of the individual datastores.

Community Treasure Hunt

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

Start Hunting!