fanqiNO1

Results 17 comments of fanqiNO1

This issue may be caused by that the version of pytorch you have is 2.1.0. We will release a pre-compiled mmcv package that supports pytorch 2.1.0 in about a week,...

> I am also facing the same issue, not able to build from source also and I am using torch version-1.11.0, even then I am facing this issue. This is...

> > > I am also facing the same issue, not able to build from source also and I am using torch version-1.11.0, even then I am facing this issue....

> I am also facing the same issue, not able to build from source also and I am using torch version-1.11.0, even then I am facing this issue. This is...

> I have the same problem. I'm trying to run mit/bevfusion, but when configuring the environment I use pip install mmcv-full==1.4.0 -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.10.1/index.html build wheel failed. My environment configuration is...

For `DataContainer`, `mmengine.BaseDataElement` has the similar function. https://github.com/open-mmlab/mmengine/blob/53474ef1ba0b166508c231fa525b55b580adf20f/mmengine/structures/base_data_element.py#L9 For collate.py, you can find the function `collate` at `mmengine.dataset.utils`. https://github.com/open-mmlab/mmengine/blob/main/mmengine/dataset/utils.py

Sorry, there is no other function which replace the collate function. However, you can refer https://github.com/open-mmlab/mmengine/blob/3b639da1ef3b46d5a42c94e3253d251d5bb6abf1/mmengine/structures/base_data_element.py#L81 to change `results_img_list.append(BaseDataElement(to_tensor(img),` to `results_img_list.append(BaseDataElement(img=to_tensor(img),`

In order to provide downstream algorithm libraries with a unified abstract data structure for data transfer between components and they need to support the wrapping of data of various natures...

It seems that it's because you use mmcv>=2.0, but the original project uses mmcv

The `data_preprocessor` was not introduced in mmcv v1.x, mmseg v0.x, it was introduced in mmcv v2.x, mmseg v1.x. If you want to use the `data_preprocessor`, please make sure mmcv>=v2.0.0, mmseg>=v1.0.0