How to use custom dataset to train custom AI model (with model zoo)

[Kuradi]

I want to use my custom dataset that is not supported by any of the .yaml files. Each yaml file example seems to only work with a dataset that is compliant to their specific use case. For example audio event detection supports ESC format and requires preprocessing configurations -or any other type of metadata- that seems to be meaningful only for .wav files; or the image classification training yaml file requires preprocessing configurations which only make sense for image data.

Can I use a custom dataset (maybe .txt or sth) to train my custom model with model zoo like in https://github.com/STMicroelectronics/stm32ai-modelzoo-services/blob/main/audio_event_detection/deployment/doc/tuto/how_to_define_and_train_my_own_model.md ? How do I tweak the user_config.yaml file in order to do so?

[LMSTM]

Hello,

If you share the use case you have in mind with us, we might be able to help you further.

If your use case isn't covered in the Model zoo, it would probably be easier to work outside of it, and write your own training code.

Your model does not need to be trained or quantized using the model zoo in order to be deployed to an STM32 device, so you can work outside of the model zoo for any or all parts of the pipeline when it is more convenient.

In case I misunderstood your question, and you wanted help with adapting your dataset to the format required by one of the use cases in the model zoo, let me know and I'll be able to help as well.

Best regards.