image classification trained on stm32-h747 - results very biased

[fanronghua0123456]

I used stm32-h747 to train image classification, I used the image classification routine, but the actual effect is very biased, my quantized_model_confusion_matrix_validation_set is as shown below, is there something wrong with my configuration?

[fanronghua0123456]

How to build a flower classifier on stm32 mcu with transfer learning (mobilenet v2) - STMicroelectronics

[Julian E.]

Hello @fanronghua0123456,

You have the same issue with this dataset you linked?

Did you try cloning the model zoo service again and setting it up again from scratch?

I cannot reproduce it, to me, it looks like and installation issue.

Have a good day,

Julian

[fanronghua0123456 · Accepted solution]

I have solved the problem, it is that my python version is wrong, and I need to install the environment strictly according to the version requirements given by ST, otherwise there will be an unknown error.