Open Vdol22 opened 7 months ago
@zen-xingle kindly ask you to answer! Is there a way to make additional training of the existing .onnx yolov8-seg models (that won't break their ability to convert to .rknn) and could you upload yolov8l-seg and x-seg .onnx versions into the zoo?
Could you please upload yolov8l-seg and x-seg .onnx versions? Or share a script to make them.
Also, is there a way to continue training existing yolo models? We need to improve detectability of 3 classes only and we have the dataset.