ImportError: cannot import name 'MultiScaleDeformableAttention' from 'mmcv.cnn.bricks.transformer'

[Coldwood1026]

The section we are using is the DETECTION section and I have executed the make.sh file in the ops folder. But we still get the following during training ImportError: cannot import name 'MultiScaleDeformableAttention' from 'mmcv.cnn.bricks.transformer' (/home/OD03/anaconda3/ envs/mmdetection/lib/python3.9/site-packages/mmcv/cnn/bricks/transformer.py) .I had a similar problem while looking for information in the mmdet forums. mmdet developers alerted MultiScaleDeformableAttention in MMCV has been moved to mmcv.ops.multi_scale_deform_attn, please update your MMCV and let's update to mmcv full1.7.1 to bypass such issues, but the platform you are developing for has mmcv support up to version 1.5.0, and the mmcv I'm using is the 1.4.2 that you recommend in the readme (our cuda version is 10.2, and our torch version is 1.12.1), and the version of your MMCV that I'm using is the one you recommend in the readme. Obviously your test run works. So I don't think it would be a problem with the mmcv version, I hope to get your answer, thanks!

The specific error is reported as follows. warnings.warn( /home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmcv/cnn/bricks/transformer.py:32: UserWarning: Fail to import MultiScaleDeformableAttention from mmcv.ops.multi_scale_deform_attn, You should install mmcv-full if you need this module. warnings.warn('Fail to import MultiScaleDeformableAttention from ' /home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmdet/models/utils/transformer.py:26: UserWarning: MultiScaleDeformableAttention in MMCV has been moved to mmcv.ops.multi_scale_deform_attn, please update your MMCV warnings.warn( Traceback (most recent call last): File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmdet/models/utils/transformer.py", line 23, in from mmcv.ops.multi_scale_deform_attn import MultiScaleDeformableAttention File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmcv/ops/init.py", line 2, in from .assign_score_withk import assign_score_withk File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmcv/ops/assign_score_withk.py", line 5, in ext_module = ext_loader.load_ext( File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmcv/utils/ext_loader.py", line 13, in load_ext ext = importlib.import_module('mmcv.' + name) File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/importlib/init.py", line 127, in import_module return _bootstrap._gcd_import(name[level:], package, level) ImportError: libtorch_cuda_cu.so: cannot open shared object file: No such file or directory

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "/home/OD03/下载/ViT-Adapter-main/detection/./train.py", line 11, in import mmdet_custom # noqa: F401,F403 File "/home/OD03/下载/ViT-Adapter-main/detection/mmdet_custom/init.py", line 2, in from .models import # noqa: F401,F403 File "/home/OD03/下载/ViT-Adapter-main/detection/mmdet_custom/models/init.py", line 2, in from .backbones import # noqa: F401,F403 File "/home/OD03/下载/ViT-Adapter-main/detection/mmdet_custom/models/backbones/init.py", line 2, in from .beit_adapter import BEiTAdapter File "/home/OD03/下载/ViT-Adapter-main/detection/mmdet_custom/models/backbones/beit_adapter.py", line 8, in from mmdet.models.builder import BACKBONES File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmdet/models/init.py", line 2, in from .backbones import * # noqa: F401,F403 File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmdet/models/backbones/init.py", line 2, in from .csp_darknet import CSPDarknet File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmdet/models/backbones/csp_darknet.py", line 11, in from ..utils import CSPLayer File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmdet/models/utils/init.py", line 17, in from .transformer import (DetrTransformerDecoder, DetrTransformerDecoderLayer, File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmdet/models/utils/transformer.py", line 29, in from mmcv.cnn.bricks.transformer import MultiScaleDeformableAttention ImportError: cannot import name 'MultiScaleDeformableAttention' from 'mmcv.cnn.bricks.transformer' (/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/mmcv/cnn/bricks/transformer.py) ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 16993) of binary: /home/OD03/anaconda3/envs/mmdetection/bin/python Traceback (most recent call last): File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/runpy.py", line 197, in _run_module_as_main return _run_code(code, main_globals, None, File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/runpy.py", line 87, in _run_code exec(code, run_globals) File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/torch/distributed/launch.py", line 193, in main() File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/torch/distributed/launch.py", line 189, in main launch(args) File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/torch/distributed/launch.py", line 174, in launch run(args) File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/torch/distributed/run.py", line 752, in run elastic_launch( File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/torch/distributed/launcher/api.py", line 131, in call return launch_agent(self._config, self._entrypoint, list(args)) File "/home/OD03/anaconda3/envs/mmdetection/lib/python3.9/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent raise ChildFailedError( torch.distributed.elastic.multiprocessing.errors.ChildFailedError:

./train.py FAILED

Failures:

[Yosugan0S0ra]

期待解决