search

Mazin-Hnewa / MS-DAYOLO

Multiscale Domain Adaptive YOLO for Cross-Domain Object Detection
MIT License
61 stars 6 forks source link

slow val #2

Open lrhjw opened 2 years ago

lrhjw commented 2 years ago

I train the data, it really use GPU. but when val, it become slow. The interface show calculation mAP. deal with 100 images with several hours. Do you know where the problem is?

Mazin-Hnewa commented 2 years ago

Are you sure that you used GPU? it should work for both training and val. try to check the Makefile and set the following: GPU=1 CUDNN=1 CUDNN_HALF=1 OPENCV=1

lrhjw commented 2 years ago

OK! ,I will Try thank you very much!