assert (mmcv_version >= digit_version(mmcv_minimum_version) AssertionError: MMCV==2.0.0 is used but incompatible. Please install mmcv>=1.3.17, <=1.7.0.

[Kenneth-X]

• after using: pip install -U openmim mim install mmengine mim install "mmcv>=2.0.0" pip install "mmsegmentation>=1.0.0"
• then using: mim download mmsegmentation --config pspnet_r50-d8_4xb2-40k_cityscapes-512x1024 --dest .

python demo/image_demo.py demo/demo.png configs/pspnet/pspnet_r50-d8_4xb2-40k_cityscapes-512x1024.py pspnet_r50-d8_512x1024_40k_cityscapes_20200605_003338-2966598c.pth --device cuda:0 --out-file result.jpg

• got error : assert (mmcv_version >= digit_version(mmcv_minimum_version) AssertionError: MMCV==2.0.0 is used but incompatible. Please install mmcv>=1.3.17, <=1.7.0.

[MatlabChen]

same problem

[apbose]

Facing the same issue while running python tools/train.py configs/convnext/convnext-small_upernet_8xb2-amp-160k_ade20k-512x512.py since it requires mmcls with mmcv compatible version mmcv>=1.4.2, <1.9.0, however the mmseg requires mmcv minimum version 2.0.0.

[shuaijun-36]

just change " mmcv_maximum_version = '1.8.0' " to "2.1.0" in the site-packages.mmdet.init.py

[wdx04]

just change " mmcv_maximum_version = '1.8.0' " to "2.1.0" in the site-packages.mmdet.init.py

That's not enough, there are many breaking changes in mmcv 2.x. The correct way may be install mmpretrain and replace all occurrence of 'mmcls' to 'mmpretrain' in the whole mmseg package.

[erencsknn]

I got a similiar error and I fixed it. You should install mmcv==2.0.0rc4 and mmcv_full==1.7.1 and pip install ftfy