How to set the confidence threshold(Conf) for YOLO Paddle inference model?

[cddddd6]

请提出你的问题 Please ask your question

Hello:
I have exported my YOLO detect model to Paddle inference model. The YOLO detect model did not give the wrong box because the confidence threshold was set to be conf=0.45. while the Paddle inference model gave the wrong box. If I lower the confidence threshold for YOLO detect model to conf=0.30, it will give the exact same wrong box as Paddle inference model. I am wondering how to set the conf for paddle model when making inference, since it is not written in any Docs. Thank you and have a good day!

Export and infer code:

from ultralytics import YOLO

Load the YOLO11 model

model = YOLO("yolo11n.pt")

Export the model to PaddlePaddle format

model.export(format="paddle") # creates '/yolo11n_paddle_model'

Load the exported PaddlePaddle model

paddle_model = YOLO("./yolo11n_paddle_model")

Run inference

results = paddle_model("https://ultralytics.com/images/bus.jpg")

[zhink]

Can you give more paddle code? We can't see how it is reasoned. Maybe refer to cfg.valid_thresh or cfg.nms_thresh in test/dygraph_to_static/yolov3.py to get some ideas.