OpenMV can't build a new firmware for Nicla Vision

Hi everyone!

I am using Nicla Vision, and have followed all steps for Image Classification with Edge Impulse and OpenMV in this link:

However, I can’t finish the last step: “7. Test the Model”. Everything is fine until I create and deploy a library with OpenMV in Edge Impulse.

Since the Nicla Vision doesn’t have any on-board SRAM we need to build the machine learning model into the firmware and load it from the flash. So, I go to the openmv girhub and fork the repository.

Then, I rename the machine learning model and the label file, and In my fork, I replace the built-in machine learning model under src/lib/libtf/models with the model downloaded from Edge Impulse. Finally, It fails to build a new firmware for Nicla Vision.

I have repeated this process 3 times, I have changed the image size to 96 and 48, used Grayscale, etc. etc and nothing works.

My openmv project on github is public, and you can see these errors below:

Note: openmv only build firmware for OPENMV2, NANO33, and ARDUINO_NANO_RP2040_CONNECT

Conclusion? This doesn’t work, or maybe I’m doing something wrong. Please I need technical support

Kind regards

If you check the build log you’ll see the compile fails because your model is too big by 640KB. So, you need to massively reduce the size of your model.

Hi @kwagyeman

Thanks for your support! I followed your advice and it works … I sacrificed some precision to get the size of my model down. I think that by trial and error I will be able to make my model more optimal.

If anyone is stuck on this issue, I’m posting a project on element14, and I’ll post my experience with OpenMV, Nicla Vision, and Edge Impulse at the link below:

Kind regards

Guillermo

1 Like

Glad you were able to get is solved.

We have a new version of the system we are about to launch. It will not have this issue.