Using a python trained neural network

31 visualizaciones (últimos 30 días)
Sania Gul
Sania Gul el 25 de Oct. de 2022
Comentada: Sania Gul el 26 de Oct. de 2022
Can you please tell me how I can save a neural network after training,in Jupiter notebook and retrieve it in MATLAB for testing on the TEST dataset.
Should I store it without .h5 or with h5 extension? A new folder "MyModel" is created, when I run generator.save("MyModel") and there is a file inside it with name "saved_model.pb"? How it is loaded and tested in MATLAB 2019a version?
  2 comentarios
Sania Gul
Sania Gul el 26 de Oct. de 2022
Wow. That's great. Tnk u sooooooooo much.
Sania Gul
Sania Gul el 26 de Oct. de 2022
MATLAB is the BEST

Iniciar sesión para comentar.

Iniciar sesión para responder a esta pregunta.

Respuesta aceptada

David Willingham
David Willingham el 25 de Oct. de 2022
Hi Sania,
I would recommend saving the model in the TensorFlow saved model format and then import it via the TensorFlow converter using the function importTensorFlowNetwork.
To save a model in the TensorFlow saved model format use the following python code:
model.save("MyModel")

Más respuestas (0)

Categorías

Más información sobre Call Python from MATLAB en Help Center y File Exchange.

Etiquetas

Productos


Versión

R2019a

Community Treasure Hunt

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

Start Hunting!

Translated by