TypeError: add(): argument 'alpha' must be Number, not NoneType #35

[reubenwenisch]

I was trying to run the repo DynamicHead and trying to run the model with a custom dataset and have the following error. In the repo the launch function is called and I get these errors.

The command I run was

DETECTRON2_DATASETS=$DATASET python train_net.py --config configs/dyhead_r50_retina_fpn_1x.yaml --num-gpus 8

Logs

Traceback (most recent call last):
  File "train_net_custom.py", line 222, in <module>
    launch(
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/detectron2/engine/launch.py", line 82, in launch
    main_func(*args)
  File "train_net_custom.py", line 216, in main
    return trainer.train()
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/detectron2/engine/defaults.py", line 484, in train
    super().train(self.start_iter, self.max_iter)
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/detectron2/engine/train_loop.py", line 149, in train
    self.run_step()
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/detectron2/engine/defaults.py", line 494, in run_step
    self._trainer.run_step()
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/detectron2/engine/train_loop.py", line 294, in run_step
    self.optimizer.step()
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/torch/optim/lr_scheduler.py", line 65, in wrapper
    return wrapped(*args, **kwargs)
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/torch/optim/optimizer.py", line 89, in wrapper
    return func(*args, **kwargs)
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/torch/optim/sgd.py", line 110, in step
    F.sgd(params_with_grad,
  File "/home/xxx/anaconda3/envs/detectron/lib/python3.8/site-packages/torch/optim/_functional.py", line 160, in sgd
    d_p = d_p.add(param, alpha=weight_decay)
TypeError: add(): argument 'alpha' must be Number, not NoneType

Detectron2 version: 0.6

Expected behaviour is the training loop runs without error.

[github-actions]

You've chosen to report an unexpected problem or bug. Unless you already know the root cause of it, please include details about it by filling the issue template.
The following information is missing: "Instructions To Reproduce the Issue and Full Logs"; "Your Environment";

[dedoogong]

I got the same error!

[reubenwenisch]

@dedoogong This error had come with using detectron2 latest version I moved it to previous version and the code started working. Hope this helps.

[awadsb1]

@reubenwenisch which version of Detectron2 did you move to? Having the same issue

[Roywangj]

I have met the same problem, i find the setting "weight_decay" don't suppport numbers like 1e-4, when i replace 1e-4 to 0.0001, it fixed.

[Yxt1212]

Same error too

[Roywangj]

这是来自QQ邮箱的假期自动回复邮件。   您好，我最近正在休假中，无法亲自回复您的邮件。我将在假期结束后，尽快给您回复。

[frostinassiky]

I have met the same problem, i find the setting "weight_decay" don't suppport numbers like 1e-4, when i replace 1e-4 to 0.0001, it fixed.

Great! The main reason is that the optimizer in this version does not handle None of weight_decay. I set it from None to 0 and my code is running smoothly.

[Roywangj]

这是来自QQ邮箱的假期自动回复邮件。   您好，我最近正在休假中，无法亲自回复您的邮件。我将在假期结束后，尽快给您回复。

[yuhao20]

I have met the same problem, i find the setting "weight_decay" don't suppport numbers like 1e-4, when i replace 1e-4 to 0.0001, it fixed.

Great! The main reason is that the optimizer in this version does not handle None of weight_decay. I set it from None to 0 and my code is running smoothly.

excuse me, could you tell me the place where we could change weight_decay? the .yaml?

[EGO-False-Sleep]

I have met the same problem, i find the setting "weight_decay" don't suppport numbers like 1e-4, when i replace 1e-4 to 0.0001, it fixed.

您好，我最近也在学习这篇开源项目，遇到了这个问题，请问您是在何处修改的weight_decay,我找到了两个可能的位置，分别是defrcn/config/defaults.py 34行 值为5e-5以及detectron2\config\defaults.py 522行值为0.0001 ，分别对应着_CC以及_C,但是在defrcn/config/config.py的get_cfg函数中，实际返回值貌似为_CC，如果您有时间，请不吝赐教，这对我非常重要，谢谢，期待您的答复

[Roywangj]

你好 这是很久以前弄的了，我记得只是从 1e-4 改成 0.0001 就可以了，具体在哪里我已经不记得了，你可以都试一下，祝好

[EGO-False-Sleep]

你好 这是很久以前弄的了，我记得只是从 1e-4 改成 0.0001 就可以了，具体在哪里我已经不记得了，你可以都试一下，祝好

感谢您的回信，我研究了很久，最后找到了一个修改方案，但是后面又陆续出现了别的报错信息，我已经打算放弃这个项目了，关于小样本目标检测任务如果您有一些好的开源项目推荐，我将万分感谢