Hello, I wanted to create a project using X-CUBE-AI and the quantized tiny yolo from model zoo (which is dedicated for deployment on STM32H747-DISCO) : https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/object_detection/pretrained_models/tiny_yolo_v2/ST_pretrainedmodel_public_dataset/coco_2017_person/tiny_yolo_v2_224
Though, when I analyze the project through CubeMX, the flash that will be used is too high, despite the model being already quantized.
Is there a proper way to reduce the flash?
I attached the .ioc file of my project.
