INTERNAL ERROR 119 during analyzing AI model

[DBany.1]

I would like to use CubeAI to run a tflite model on microcontroller, but I get ERROR 119 message during the analyzing model step. I tried to search for this error code, but could not find any sources that reveals what it actually means. I attach the screenshot of the message. The log file mentioned in the error message provides no further info about the problem.

CubeMX and CubeAI is runnining in a Windows environment, and has been updated to the latest versions.

The tflite model works in other environments without any problem. It has been converted from a tensorflow keras model, using GRU and Dense layers. Could it be the problem, that CubeAI does not support tflite models that has been converted from GRU layers? If so, is there any specification what kind of keras layers can be used?

Thanks,

Daniel

[jean-michel.d]

​Hello Daniel,

INTERNAL ERROR 119 seems indicate an error raised by the TFlite importer, not clearly managed by the tools.

What is the version of TF.keras used to generate the model?

According the doc (https://www.tensorflow.org/lite/guide/ops_compatibility), GRU seems not supported! But as your model works, is it possible to share the used model? On our side the GRU/LSTM operators are not supported by the TFLite imported but with the Keras imported.

Have you try to to import the original TF Keras model?

Regards,

Jean-Michel

[DBany.1]

Hi Jean-Michel,

first thanks for the quick reply!

I tried the Keras model, and it works without any problem :)

Daniel