-
Notifications
You must be signed in to change notification settings - Fork 51
Open
Description
https://github.com/cuiziteng/Illumination-Adaptive-Transformer/blob/main/IAT_high/IAT_mmdetection/mmdet/models/detectors/IAT_detector/IAT_yolo.py
extract_feat函数和forward_train只初始化在init函数里面,实际上是不能够被反向传播的,我在extract_feat加了一行打印函数,训练过程中并没有打印。将extract_feat函数和forward_train移出init函数,作为IAT_yolo的成员函数,实际上是重写SingleStageDetecto的函数,这样extract_feat才会参与反向传播,打印日志。或者是我理解错了,请作者耐心解答一下。
Metadata
Metadata
Assignees
Labels
No labels