MMClassification Release V0.14.0

Highlights

• Add transformer-in-transformer backbone and pretrain checkpoints, refers to the paper.
• Add Chinese colab tutorial.
• Provide dockerfile to build mmcls dev docker image.

New Features

• Add transformer in transformer backbone and pretrain checkpoints. (#339)
• Support mim, welcome to use mim to manage your mmcls project. (#376)
• Add Dockerfile. (#365)
• Add ResNeSt configs. (#332)

Improvements

• Use the presistent_works option if available, to accelerate training. (#349)
• Add Chinese ipynb tutorial. (#306)
• Refactor unit tests. (#321)
• Support to test mmdet inference with mmcls backbone. (#343)
• Use zero as default value of thrs in metrics. (#341)

Bug Fixes

• Fix ImageNet dataset annotation file parse bug. (#370)
• Fix docstring typo and init bug in ShuffleNetV1. (#374)
• Use local ATTENTION registry to avoid conflict with other repositories. (#376)
• Fix swin transformer config bug. (#355)
• Fix patch_cfg argument bug in SwinTransformer. (#368)
• Fix duplicate init_weights call in ViT init function. (#373)
• Fix broken _base_ link in a resnet config. (#361)
• Fix vgg-19 model link missing. (#363)

Contributors

A total of 8 developers contributed to this release.

@Ezra-Yu, @HIT-cwh, @Junjun2016, @LXXXXR, @mzr1996, @pvys, @wangruohui, @ZwwWayne