TDA4VH-Q1: Demo running batch on the board

cheng han

Part Number: TDA4VH-Q1

HI TI:

I noticed that TI’s official website has released the RTOS 9.0 SDK update, and in the release notes, it mentioned that version 9.0 supports batch processing across multiple C7X cores.

However, I couldn’t find any related demos in the RTOS 9.0 SDK.

Can you provide some guidance on how to use batch processing? Could you also provide a demo, please?

Thank you for your support.

over 3 years ago

0 Anand Pathak over 3 years ago

TI__Genius 9065 points

Batch processing feature is taken care of internally as part of TIDL execution, so you just need to export models to have batch dimension of 4 and can pass the model to any of the existing demos. Please refer to the following steps for additional compilation options required to compile the model to be run using batch processing mode : https://github.com/TexasInstruments/edgeai-tidl-tools/blob/master/examples/osrt_python/README.md#options-for-devices-with-multiple-dsp-cores

Please continue the thread in case of any issues in running above.

Regards,

Anand

Processors

Processors forum

TDA4VH-Q1: Demo running batch on the board