qq184861643
qq184861643
I dont know why inserted code is arranged like that, sorry
I've figured out that the new architecture would lead the outputs larger than the inputs with few order of magnitudes, but I still dont know why...
@gaoxiang12 @DamonMIN actually the modified g2o's default output directory is into /ORBSLAM2_with_pointcloud_map-master/g2o_with_orbslam2/lib, I've inspected it via ccmake.However I also have the same lib into my usr/local/lib somehow,so I don't know...
@hcx1231 @KevinQian97 @guoruoqian Hi guys, I've got the same problem. Have you guys solved it? Could you plz tell me how to solve it?
@jwyang @doronpor Hi，if any of you already solve this problem in your own code, could you please kindly modify the code on this github? I've got some trouble to solve...
@Feynman27 @doronpor Hi，and I have an additional question if you may allow me to ask. It seems that you guys assume that rpn_cls_score and rpn_bbox_pred have a exactly arranged order...
> Resuming training works well for me. I didnt change anything in train method and use Arcface head. @boomberung thx. Then maybe my problem lies in nn.DataParallel. I will try...
@boomberung Hi! have you figured it out how to solve this? I've tried several methods but still can't fix it
@DecentMakeover if we can't solve the resuming issue I don't think fine-tuning is possible
@TheShadow29 @yhenon Hi, sorry for @ you guys but I have a question in code and I think it maybe related to this issue. At model.py class ResNet, in its...