Hello!
I was able to create yolov5n_int8_fuse.cvimodel by following the steps on the following site.
I then copied it with the sdk to a Duo256M board and ran it, and I get the following results. (left: original, right: result)
When I run model_deploy.py, I specify --chip cv180x
, but I have a feeling that it is different because Duo256M is SG2002. I looked at the source, but didn’t see any models that might be applicable.
Also, I copied the above model to Duo (64M) and it seems to detect correctly as follows
I would like to know if the “TPU” section of the following URL is for Duo and not for Duo256M.
Thanks!