BangguWu / ECANet

Code for ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks
MIT License
1.24k stars 197 forks source link

Info on configuration on Faster-rcnn ECA #59

Open Yashay opened 3 years ago

Yashay commented 3 years ago

Hi,

I hope you are well. Great work!

I assume that you used the mmdetection framework for the detection task. What training schedule did you use 1x or 2x? And for Faster-RCNN SE-NET did you train from scratch?

Kind Regards Yashay

BangguWu commented 3 years ago

I used 1x training schedule, and Faster-RCNN SE-NET is trained from scratch