Confidence scores are too low despite while using deepfashion2 validation set

[carlosalinas]

I've used the YOLOv3 model for Deepfashion2 with the validation set given by deepfashion2 and my scores are so low despite using 5,000 and 10,000 images. The mean average precision is low as well. Can I ask how did you come up with your own result for coco evaluation?

[simaiden]

Hi, sorry for the late reply

Hoy did you calculate the average precision? I used darknet to get the .json file with detections in coco format, but you can get your owns as well. If you get the .json file manually, you have to set the confidence threshold as lower as possible (i.e. 0.001)