PyTorch-YOLOv3
PyTorch-YOLOv3 copied to clipboard
Inference Something Trouble lead to slow
I watch your code in evaluate. I find your NMS process all proposals bbox. It's make some trouble in early epoch due to some many remained proposals. So, I Suggest in here that set remain paras in NMS likely MaskRcnn-BenchMark, for examle 100 remained. in a whole, good job.