Openmixup

New Features

- Support various popular backbones (ConvNets and ViTs) and update config files.
- Support various handcrafted methods and optimization-based methods (e.g., [PuzzleMix](https://arxiv.org/abs/2009.06962), [AutoMix](https://arxiv.org/pdf/2103.13027), [SAMix](https://arxiv.org/pdf/2111.15454), [DecoupleMix](https://arxiv.org/abs/2203.10761), etc.). Config files generation of mixup methods are supported.
- Provide supervised image classification benchmarks in model_zoo and results (on updating).

Bug Fixes

- Fix bugs of new mixup methods (e.g., gco for Puzzlemix, etc.).

New Features

- Support various popular backbones (popular ConvNets and ViTs).
- Support mixed precision training (NVIDIA Apex or MMCV Apex).
- Support supervised, self- & semi-supervised learning methods and benchmarks.
- Support fast configs generation from a basic config file by `auto_train.py`.

Bug Fixes

- Fix bugs of code refactoring (backbones, fp16 training, etc.).

OpenSelfSup (v0.3.0, 14/10/2020) Supported Features

This repo is originally built on OpenSelfSup (the old version of [MMSelfSup](https://github.com/open-mmlab/mmselfsup)) and borrows some implementations from [MMClassification](https://github.com/open-mmlab/mmclassification).

- Mixed Precision Training (based on NVIDIA Apex for **PyTorch 1.6**).
- Improvement of GaussianBlur doubles the training speed of MoCo V2, SimCLR, and BYOL.
- More benchmarking results, including benchmarks on Places, VOC, COCO, and linear/semi-supervised benchmarks.
- Fix bugs in moco v2 and BYOL so that the reported results are reproducible.
- Provide benchmarking results and model download links.
- Support updating the network every several iterations (accumulation).
- Support LARS and LAMB optimizer with Nesterov (LAMB from [MMClassification](https://github.com/open-mmlab/mmclassification)).
- Support excluding specific parameter-wise settings from the optimizer updating.