search

megvii-model / YOLOF

MIT License
809 stars 115 forks source link

about accuracy #50

Open shenhaibb opened 2 years ago

shenhaibb commented 2 years ago

I use res50-C5 when I don't use the pre-training model, the accuracy is only 59%. After using the pre-training model, the accuracy reaches 80%. However, my experiment on YOLOv5s is only 75%. Should I use the pre-training model? What may be the reason for my situation?

ytoon commented 2 years ago

Yes, it is better to use the pre-trained model given performance and training time. We suggest you to read this paper for a deeper understanding.