I see that NXP has this article about running NPU model on external with eiq tflm_label_image_ext_mem example memory https://docs.nxp.com/bundle/AN14700/page/topics/external_memory.html
It only mentioned about eiq example, but did not mention details how to modify this example for new NPU models on external memory. In opposite, NXP have very detailed guideline how to modify source code to run new NPU models on SRAM https://docs.nxp.com/bundle/AN14700/page/topics/Note_about_the_source_code_in_the_document.html
Currently, I want to deploy NPU model on external memory. If you have detailed guideline, please share with us.
Thank you.
Hi @nnxxpp ,
Thank you so much for your interest in our products and for using our community.
If you want to run the NPU model from external memory, I suggest starting from the tflm_label_image_ext_mem example instead of the normal SRAM-based tflm_label_image example.
Please refer to the tflm_label_image_ext_mem MCUXpresso SDK documentation for more details.
tflm_label_image_ext_mem — MCUXpresso SDK Documentation
Wish it helps you
Best Regards
May
Hi @nnxxpp ,
Thank you for your feedback. I'm glad the documentation was helpful.
If you have any new questions in the future, please feel free to post them in the NXP Community.
We'll be happy to help.
Best Regards
May
Thank you. I have tried several trials and found that we must to not use sequencer when locate model on external memory same as you mentioned.
This is the first time I see NXP's documentation that you attached. Thank you so much for helpful documentation.