humu789

Results 19 comments of humu789

We will release quantization in Q4 2022, if you have some requirements and advice, welcome to tell us.

Thanks. You can refer to https://mmrazor.readthedocs.io/en/main/user_guides/quantization_user_guide.html for basic usage and examples. For openmmlab's models, it is theoretically possible to apply qat to your model by changing the config file.

OpenMMLab repos all upgraded, you can upgrade mmcls, thus the requirements for mmcv are same. You can refer to https://github.com/open-mmlab/mmpretrain/tree/mmcls-1.x

Sorry for the late reply because of health reasons. `quantize` branch is WIP now, we will update it with more complete functions and docs as soon.

sorry for the late reply, you should use the old version: original `master` branch, the old version is no longer maintained, please use the `main` branch that is mmrazor 1.0

Is your model based on openmmlab repos, or custom model?

Sorry for the late reply, we will release the quantization part at the end of the year if all goes well. The transformer structure will not be supported in the...

> > Sorry for the late reply, we will release the quantization part at the end of the year if all goes well. The transformer structure will not be supported...