some error with fvcore

[xiaowanzizz]

File "/home/zzf/miniconda3/envs/torch1.7.1/lib/python3.7/site-packages/fvcore-0.1.3.post20210317-py3.7.egg/fvcore/nn/giou_loss.py", line 32, in giou_loss
AssertionError: bad box: x1 larger than x2

[chensnathan]

Could you provide more details about the command you used and post the training log?

[xiaowanzizz]

Exception during training:
Traceback (most recent call last):
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/train_loop.py", line 138, in train
self.run_step()
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/defaults.py", line 441, in run_step
self._trainer.run_step()
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/train_loop.py", line 232, in run_step
loss_dict = self.model(data)
File "/home/zzf/miniconda3/envs/torch1.7.1/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/zzf/Desktop/detectron2-0.4/YOLOF-master/yolof/modeling/yolof.py", line 295, in forward
pred_logits, pred_anchor_deltas)
File "/home/zzf/Desktop/detectron2-0.4/YOLOF-master/yolof/modeling/yolof.py", line 407, in losses
matched_predicted_boxes, target_boxes, reduction="sum")
File "/home/zzf/miniconda3/envs/torch1.7.1/lib/python3.7/site-packages/fvcore-0.1.3.post20210317-py3.7.egg/fvcore/nn/giou_loss.py", line 32, in giou_loss
assert (x2 >= x1).all(), "bad box: x1 larger than x2"

Traceback (most recent call last):
File "./tools/train_net.py", line 249, in
args=(args,),
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/launch.py", line 62, in launch
main_func(*args)
File "./tools/train_net.py", line 236, in main
return trainer.train()
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/defaults.py", line 431, in train
super().train(self.start_iter, self.max_iter)
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/train_loop.py", line 138, in train
self.run_step()
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/defaults.py", line 441, in run_step
self._trainer.run_step()
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/defaults.py", line 441, in run_step
self._trainer.run_step()
File "/home/zzf/Desktop/CenterNet2/detectron2/engine/train_loop.py", line 232, in run_step
loss_dict = self.model(data)
File "/home/zzf/miniconda3/envs/torch1.7.1/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/zzf/Desktop/detectron2-0.4/YOLOF-master/yolof/modeling/yolof.py", line 295, in forward
pred_logits, pred_anchor_deltas)
File "/home/zzf/Desktop/detectron2-0.4/YOLOF-master/yolof/modeling/yolof.py", line 407, in losses
matched_predicted_boxes, target_boxes, reduction="sum")
File "/home/zzf/miniconda3/envs/torch1.7.1/lib/python3.7/site-packages/fvcore-0.1.3.post20210317-py3.7.egg/fvcore/nn/giou_loss.py", line 32, in giou_loss
AssertionError: bad box: x1 larger than x2

python ./tools/train_net.py --num-gpus 1 --config-file ./configs/yolof_R_50_C5_1x.yaml --> command

[chensnathan]

The learning rate and steps are set for 8 GPUS. Have you modified the learning rate and steps in the config file according to the linear learning rate scaling rule as said in Detectron2?

[xiaowanzizz]

thanks. changing the learning rate can solve this problem.

[yarkable]

meet same problem now……， really awful

[DLGreenhand]

The learning rate and steps are set for 8 GPUS. Have you modified the learning rate and steps in the config file according to the linear learning rate scaling rule as said in Detectron2?

Thanks！