zimenglan
zimenglan
hi @gen-ko how to get the real shift for very complex model?
hi @DHCZ, thanks
btw, the training speed of `configs/hrnet/mask_rcnn_hrnet_w18_1x.yaml` is too slow, compared with `/configs/e2e_mask_rcnn_R_50_FPN_1x.yaml` using single gpu with 2 images per iteration (time: 0.5182 (0.4975) v.s. time: 0.3253 (0.3393)).
hi @mshmoon do u solve this problem?
hi @luxiin can u show me the code of it, since i can't find it. thanks.
thanks @luxiin btw, in mmdetection, it seems to use reg-loss in the [config file ](https://github.com/open-mmlab/mmdetection/blob/master/configs/grid_rcnn/grid_rcnn_gn_head_r50_fpn_2x.py#L35) and in the [detectors/grid_rcnn.py](https://github.com/open-mmlab/mmdetection/blob/master/mmdet/models/detectors/grid_rcnn.py#L146).
u are right. thanks.
hi @kis2u any update about the training code and the model? thanks
+1, architecture of GE-theta plus is more powful.