KeyError: 'Mask2FormerVideoCustomMatching is not in the models registry'

Traceback (most recent call last):
File "/home/wm/Tube-Link/tools/train.py", line 250, in
main()
File "/home/wm/Tube-Link/tools/train.py", line 220, in main
model = build_detector(
File "/mnt/12T/wm/conda/bin/turbe0/lib/python3.9/site-packages/mmdet/models/builder.py", line 58, in build_detector
return DETECTORS.build(
File "/mnt/12T/wm/conda/bin/turbe0/lib/python3.9/site-packages/mmcv/utils/registry.py", line 238, in build
return self.build_func(*args, **kwargs, registry=self)
File "/mnt/12T/wm/conda/bin/turbe0/lib/python3.9/site-packages/mmcv/cnn/builder.py", line 27, in build_model_from_cfg
return build_from_cfg(cfg, registry, default_args)
File "/mnt/12T/wm/conda/bin/turbe0/lib/python3.9/site-packages/mmcv/utils/registry.py", line 62, in build_from_cfg
raise KeyError(
KeyError: 'Mask2FormerVideoCustomMatching is not in the models registry'.

(/mnt/12T/wm/conda/bin/turbe0) wm@RTX6000Server2:~/Tube-Link$ python
Python 3.9.12 (main, Jun 1 2022, 11:38:51)
[GCC 7.5.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.

import mmdet
/mnt/12T/wm/conda/bin/turbe0/lib/python3.9/site-packages/mmcv/init.py:20: UserWarning: On January 1, 2023, MMCV will release v2.0.0, in which it will remove components related to the training process and add a data transformation module. In addition, it will rename the package names mmcv to mmcv-lite and mmcv-full to mmcv. See https://github.com/open-mmlab/mmcv/blob/master/docs/en/compatibility.md for more details.
warnings.warn(
mmdet.version
'2.28.2'
import mmcv
mmcv.version
'1.7.2'

(/mnt/12T/wm/conda/bin/turbe0) wm@RTX6000Server2:/Tube-Link$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Tue_Feb__7_19:32:13_PST_2023
Cuda compilation tools, release 12.1, V12.1.66
Build cuda_12.1.r12.1/compiler.32415258_0
(/mnt/12T/wm/conda/bin/turbe0) wm@RTX6000Server2:/Tube-Link$ python
Python 3.9.12 (main, Jun 1 2022, 11:38:51)
[GCC 7.5.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.

import torch
torch.version
'2.1.0+cu121'

Start at tools/train.py around lines 220 and 250, where build_detector is called, and inspect how the Mask2FormerVideoCustomMatching model is configured and imported. Check the reported mmdet 2.28.2, mmcv 1.7.2, Python 3.9.12, and CUDA 12.1 environment while reproducing the registry error. Done means the configured model is available in the registry and the training entry point proceeds past model construction.