[Feature Request] Apex support for mixed precision training

[nijkah]

Describe the feature

Motivation
Many recent research works are based on apex for mixed precision training.

So I want to add this feature to mmcv with the name of class EpochBasedRunnerApex.
I am thinking implementation of just wrapping EpochBasedRunner as in Swin-Transformer Implementation.

Is it good idea to mmcv? Then I will create the PR.

Related resources
https://github.com/SwinTransformer/Swin-Transformer-Object-Detection
https://github.com/facebookresearch/ConvNeXt
https://github.com/microsoft/unilm/tree/master/beit/semantic_segmentation

[HAOCHENYE]

Thanks for your advice, but what is the advantage of apex over torch.cuda.amp? mmcv implements mixed precision training via Fp16OptimizerHook.

[nijkah]

@HAOCHENYE
From this link, it says Apex is a first implementation of amp, and deprecated.
So, it doesn't seem to need to support that.
I think I have to check why these works used apex for amp.