Jingwei Zhang
Jingwei Zhang
Maybe you should also add some code in init function: ` if self.image_backend is not None: mmcv.use_backend(self.image_backend) ` if not, you can not use `turbojpeg` as backend. [jpeg](https://github.com/open-mmlab/mmcv/blob/master/mmcv/image/io.py#L241) However, now,...
Please complete all unit tests and docstrings and make sure all unit tests pass. You can refer to the implementation of STARK.
> > Please complete all unit tests and docstrings and make sure all unit tests pass. You can refer to the implementation of STARK. > > In unit tests, `run...
> > > > Please complete all unit tests and docstrings and make sure all unit tests pass. You can refer to the implementation of STARK. > > > >...
Please complete all unit tests and docstrings and make sure all unit tests pass. You can refer to the implementation of STARK. Also, I noticed that this PR uses lots...
In order to facilitate later maintenance and secondary development, please add unit tests of forward for these important functions or classes: `MixformerHead`, `MixFormer`, `MixFormerBackbone`. You can refer to SiamRPN and...
All the UTs run successfully on the local computer. LGTM @GT9505
Thanks for your feedback!
We have not supported it yet. We'll appreciate it if you can add such new features to our codebase.
Are you sure the ground truth boxes loaded from the file are in the right format? You can check it in `prepare_data` function in `CocoVideoDataset`.