error with cascade_rcnn_mdconv_c3-c5_x101_64x4d_fpn_1x_0.py

[chituma110]

Thanks for your bug report and we appreciate that a lot.

Checklist

• I have searched related issues but could not get the expected help.
• The bug has not been fixed in the latest version.

Describe the bug
RuntimeError: While copying the parameter named layer2.0.conv2.weight, whose dimensions in the model are torch.Size([512, 512, 3, 3]) and whose dimensions in the checkpoint are torch.Size([512, 8, 3, 3]).

Error traceback

1. What command or script did you run?
   /home/mmdetection/configs/cascade_rcnn_mdconv_c3-c5_x101_64x4d_fpn_1x_0.py
   --validate
   --gpus
   8
   --launcher
   pytorch
   Environment

• OS: [e.g., Ubuntu 16.04.6]
• GCC [e.g., 5.4.0]
• PyTorch version [e.g., 1.1.0]
• How you installed PyTorch [ conda]
• GPU model [e.g., TitanXp]
• CUDA 10.0

[hellock]

The command seems not complete. Could you provide the running command and config file (if you modified it)?

[chituma110]

The full completed command is
python /home/mmdetection/tools/train.py /home/mmdetection/configs/cascade_rcnn_mdconv_c3-c5_x101_64x4d_fpn_1x_0.py
--validate
--gpus
8
--launcher
pytorch

And I got the following error

RuntimeError: While copying the parameter named layer2.0.conv2.weight, whose dimensions in the model are torch.Size([512, 512, 3, 3]) and whose dimensions in the checkpoint are torch.Size([512, 8, 3, 3]).

The base net is resnext101 which is downloaded from
'resnext101_64x4d': 'https://s3.ap-northeast-2.amazonaws.com/open-mmlab/pretrain/third_party/resnext101_64x4d-ee2c6f71.pth', # noqa: E501

[hellock]

I cannot even run with the provided command above (other errors were raised before loading checkpoints), are you sure about that?
BTW, have you checked the second item in the checklist?