Keras 2.3.1 models

9 views (last 30 days)
Larry Anderson
Larry Anderson on 26 May 2020
Is there a way to import Keras 2.3.1 models into Matlab? I have the Keras Importer from the File Exchange installed but looks like it will only import up to version 2.2.4.

Answers (2)

Sivylla Paraskevopoulou
Sivylla Paraskevopoulou on 9 May 2022
Edited: Sivylla Paraskevopoulou on 9 May 2022
The importKerasNetwork function offers limited support for TensorFlow-Keras versions 2.2.5 to 2.4.0. For more information, see Limitations.
It is recommended that you use importTensorFlowNetwork to import your model. The importTensorFlowNetwork function supports TensorFlow versions v2.0 to 2.6. Moreover, importTensorFlowNetwork tries to generate a custom layer when you import a custom TensorFlow layer or when the software cannot convert a TensorFlow layer into an equivalent built-in MATLAB layer.
Note that you must first convert your model to the SavedModel format before importing it by using importTensorFlowNetwork.
  2 Comments
Juan David Diaz
Juan David Diaz on 7 Dec 2022
The importTensorFlowNetwork function is not included in the support package.
Sivylla Paraskevopoulou
Sivylla Paraskevopoulou on 7 Dec 2022
The importTensorFlowNetwork function was introduced to the Deep Learning Toolbox Converter for TensorFlow Models support package in R2021a. Are you using MATLAB R2021a or newer?

Sign in to comment.


Srivardhan Gadila
Srivardhan Gadila on 30 May 2020
As per my knowledge the "Deep Learning Toolbox Importer for Keras Models support package" installed from Add-On Explorer supports upto Keras versions 2.2.4.
Still you will be able to import a keras model which is saved in versions newer than 2.2.4 but with the following the warning:
Warning: File 'CNNKeras.h5' was saved in Keras version '2.3.1'. Import of Keras versions newer than '2.2.4' is not yet supported.
The imported model may not exactly match the model saved in the Keras file.
You can refer to Import and Export Networks for the available functions.

Products


Release

R2020a

Community Treasure Hunt

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

Start Hunting!