helq2612

Hi, I have trained the voc-> clipart with the default setting (8 GPUs, default configurations), and the loss gets nan after iter=21k (the SEMISUPNET.DIS_LOSS_WEIGHT is set to 0.1 as default)....

Training got diverged after 25K iterations (with SEMISUPNET.DIS_LOSS_WEIGHT=0.05). The best performance is AP50=36.3788. The error message is here: ``` File "/project/codes/adaptive_teacher/adapteacher/modeling/proposal_generator/rpn.py", line 53, in forward -- anchors, pred_objectness_logits, pred_anchor_deltas, images.image_sizes...

Hi @michaelku1 , are you using MAX_SIZE_TRAIN=1200, MIN_SIZE_TRAIN=(600,), or the default settings in the config file (see discussion here https://github.com/facebookresearch/adaptive_teacher/issues/23#issue-1291276957)? Could you share your config file and the bash script?...

Thank you! @yujheli I am training on it now with the updated files. ``` CUDA_VISIBLE_DEVICES=1,2,3,4 python -W ignore train_net.py --num-gpus 4 --config configs/faster_rcnn_R101_cross_clipart.yaml OUTPUT_DIR output/exp_clipart SOLVER.IMG_PER_BATCH_LABEL 16 SOLVER.IMG_PER_BATCH_UNLABEL 16 ```...

Sorry to bother you again, @yujheli . But the training is still diverged. ``` [07/11 16:00:47 d2.utils.events]: eta: 1 day, 20:50:28 iter: 27618 total_loss: 2.31 loss_cls: 0.09012 loss_box_reg: 0.147 loss_rpn_cls:...

I think the temp=20 currently is used in the image position encoding only, see [main.py](https://github.com/IDEA-opensource/DAB-DETR/blob/309f6ad92af7a62d7732c1bdf1e0c7a69a7bdaef/main.py#L59) and [position_encoding.py](https://github.com/IDEA-opensource/DAB-DETR/blob/309f6ad92af7a62d7732c1bdf1e0c7a69a7bdaef/models/DAB_DETR/position_encoding.py#L66). Agree with @[YellowPig-zp](https://github.com/YellowPig-zp), this temp=20 should be also applied to the box position...

@lxtGH What do you mean about this threshold? Do you mean the threshold to get the binary mask? If it is, it is here in misc.py: https://github.com/Mhaiyang/ICCV2019_MirrorNet/blob/032bc60dbd257a35871d68773ceaef9cfd445e05/misc.py#L133 If it means...

Hi @lxtGH , could you reproduce the R3Net result in Table 1? What is the performance you got?

vote for "Semantic Seg/Instance Seg/Key points tasks" Also hope this framework can support other backbones, e.g. ViT

我也是, 我感觉在validation上差不多, 但是在官网上跑的结果就差一些, 不知道是不是还有一些设置需要微调什么的.