Initial commit

docs/en/notes/changelog_v2.x.md

+# Changelog v2.x
+
+### v2.25.0 (31/5/2022)
+
+#### Highlights
+
+- Support dedicated `WandbLogger` hook
+- Support [ConvNeXt](configs/convnext), [DDOD](configs/ddod), [SOLOv2](configs/solov2)
+- Support [Mask2Former](configs/mask2former) for instance segmentation
+- Rename [config files of Mask2Former](configs/mask2former)
+
+#### Backwards incompatible changes
+
+- Rename [config files of Mask2Former](configs/mask2former) (#7571)
+
+  <table align="center">
+    <thead>
+        <tr align='center'>
+            <td>before v2.25.0</td>
+            <td>after v2.25.0</td>
+        </tr>
+    </thead>
+    <tbody><tr valign='top'>
+    <th>
+
+  - `mask2former_xxx_coco.py` represents config files for **panoptic segmentation**.
+
+  </th>
+    <th>
+
+  - `mask2former_xxx_coco.py` represents config files for **instance segmentation**.
+  - `mask2former_xxx_coco-panoptic.py` represents config files for **panoptic segmentation**.
+
+  </th></tr>
+  </tbody></table>
+
+#### New Features
+
+- Support [ConvNeXt](https://arxiv.org/abs/2201.03545) (#7281)
+- Support [DDOD](https://arxiv.org/abs/2107.02963) (#7279)
+- Support [SOLOv2](https://arxiv.org/abs/2003.10152) (#7441)
+- Support [Mask2Former](https://arxiv.org/abs/2112.01527) for instance segmentation (#7571, #8032)
+
+#### Bug Fixes
+
+- Enable YOLOX training on different devices (#7912)
+- Fix the log plot error when evaluation with `interval != 1` (#7784)
+- Fix RuntimeError of HTC (#8083)
+
+#### Improvements
+
+- Support dedicated `WandbLogger` hook (#7459)
+
+  Users can set
+
+  ```python
+  cfg.log_config.hooks = [
+    dict(type='MMDetWandbHook',
+         init_kwargs={'project': 'MMDetection-tutorial'},
+         interval=10,
+         log_checkpoint=True,
+         log_checkpoint_metadata=True,
+         num_eval_images=10)]
+  ```
+
+  in the config to use `MMDetWandbHook`. Example can be found in this [colab tutorial](https://colab.research.google.com/drive/1RCSXHZwDZvakFh3eo9RuNrJbCGqD0dru?usp=sharing#scrollTo=WTEdPDRaBz2C)
+
+- Add `AvoidOOM` to avoid OOM (#7434, #8091)
+
+  Try to use `AvoidCUDAOOM` to avoid GPU out of memory. It will first retry after calling `torch.cuda.empty_cache()`. If it still fails, it will then retry by converting the type of inputs to FP16 format. If it still fails, it will try to copy inputs from GPUs to CPUs to continue computing. Try AvoidOOM in code to make the code continue to run when GPU memory runs out:
+
+  ```python
+  from mmdet.utils import AvoidCUDAOOM
+
+  output = AvoidCUDAOOM.retry_if_cuda_oom(some_function)(input1, input2)
+  ```
+
+  Users can also try `AvoidCUDAOOM` as a decorator to make the code continue to run when GPU memory runs out:
+
+  ```python
+  from mmdet.utils import AvoidCUDAOOM
+
+  @AvoidCUDAOOM.retry_if_cuda_oom
+  def function(*args, **kwargs):
+      ...
+      return xxx
+  ```
+
+- Support reading `gpu_collect` from `cfg.evaluation.gpu_collect` (#7672)
+
+- Speedup the Video Inference by Accelerating data-loading Stage (#7832)
+
+- Support replacing the `${key}` with the value of `cfg.key` (#7492)
+
+- Accelerate result analysis in `analyze_result.py`. The evaluation time is speedup by 10 ~ 15 times and only tasks 10 ~ 15 minutes now. (#7891)
+
+- Support to set `block_dilations` in `DilatedEncoder` (#7812)
+
+- Support panoptic segmentation result analysis (#7922)
+
+- Release DyHead with Swin-Large backbone (#7733)
+
+- Documentations updating and adding
+
+  - Fix wrong default type of `act_cfg` in `SwinTransformer` (#7794)
+  - Fix text errors in the tutorials (#7959)
+  - Rewrite the [installation guide](docs/en/get_started.md) (#7897)
+  - [Useful hooks](docs/en/tutorials/useful_hooks.md) (#7810)
+  - Fix heading anchor in documentation  (#8006)
+  - Replace `markdownlint` with `mdformat` for avoiding installing ruby (#8009)
+
+#### Contributors
+
+A total of 20 developers contributed to this release.
+
+Thanks @ZwwWayne, @DarthThomas, @solyaH, @LutingWang, @chenxinfeng4, @Czm369, @Chenastron, @chhluo, @austinmw, @Shanyaliux @hellock, @Y-M-Y, @jbwang1997, @hhaAndroid, @Irvingao, @zhanggefan, @BIGWangYuDong, @Keiku, @PeterVennerstrom, @ayulockin
+
+### v2.24.0 (26/4/2022)
+
+#### Highlights
+
+- Support [Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation](https://arxiv.org/abs/2012.07177)
+- Support automatically scaling LR according to GPU number and samples per GPU
+- Support Class Aware Sampler that improves performance on OpenImages Dataset
+
+#### New Features
+
+- Support [Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation](https://arxiv.org/abs/2012.07177), see [example configs](configs/simple_copy_paste/mask_rcnn_r50_fpn_syncbn-all_rpn-2conv_ssj_scp_32x2_270k_coco.py) (#7501)
+
+- Support Class Aware Sampler, users can set
+
+  ```python
+  data=dict(train_dataloader=dict(class_aware_sampler=dict(num_sample_class=1))))
+  ```
+
+  in the config to use `ClassAwareSampler`. Examples can be found in [the configs of OpenImages Dataset](https://github.com/open-mmlab/mmdetection/tree/main/configs/openimages/faster_rcnn_r50_fpn_32x2_cas_1x_openimages.py).  (#7436)
+
+- Support automatically scaling LR according to GPU number and samples per GPU. (#7482)
+  In each config, there is a corresponding config of auto-scaling LR as below,
+
+  ```python
+  auto_scale_lr = dict(enable=True, base_batch_size=N)
+  ```
+
+  where `N` is the batch size used for the current learning rate in the config (also equals to `samples_per_gpu` * gpu number to train this config).
+  By default, we set `enable=False` so that the original usages will not be affected. Users can set `enable=True` in each config or add `--auto-scale-lr` after the command line to enable this feature and should check the correctness of `base_batch_size` in customized configs.
+
+- Support setting dataloader arguments in config and add functions to handle config compatibility. (#7668)
+  The comparison between the old and new usages is as below.
+
+  <table align="center">
+    <thead>
+        <tr align='center'>
+            <td>v2.23.0</td>
+            <td>v2.24.0</td>
+        </tr>
+    </thead>
+    <tbody><tr valign='top'>
+    <th>
+
+  ```python
+  data = dict(
+      samples_per_gpu=64, workers_per_gpu=4,
+      train=dict(type='xxx', ...),
+      val=dict(type='xxx', samples_per_gpu=4, ...),
+      test=dict(type='xxx', ...),
+  )
+  ```
+
+  </th>
+    <th>
+
+  ```python
+  # A recommended config that is clear
+  data = dict(
+      train=dict(type='xxx', ...),
+      val=dict(type='xxx', ...),
+      test=dict(type='xxx', ...),
+      # Use different batch size during inference.
+      train_dataloader=dict(samples_per_gpu=64, workers_per_gpu=4),
+      val_dataloader=dict(samples_per_gpu=8, workers_per_gpu=2),
+      test_dataloader=dict(samples_per_gpu=8, workers_per_gpu=2),
+  )
+
+  # Old style still works but allows to set more arguments about data loaders
+  data = dict(
+      samples_per_gpu=64,  # only works for train_dataloader
+      workers_per_gpu=4,  # only works for train_dataloader
+      train=dict(type='xxx', ...),
+      val=dict(type='xxx', ...),
+      test=dict(type='xxx', ...),
+      # Use different batch size during inference.
+      val_dataloader=dict(samples_per_gpu=8, workers_per_gpu=2),
+      test_dataloader=dict(samples_per_gpu=8, workers_per_gpu=2),
+  )
+  ```
+
+  </th></tr>
+  </tbody></table>
+
+- Support memory profile hook. Users can use it to monitor the memory usages during training as below (#7560)
+
+  ```python
+  custom_hooks = [
+      dict(type='MemoryProfilerHook', interval=50)
+  ]
+  ```
+
+- Support to run on PyTorch with MLU chip (#7578)
+
+- Support re-spliting data batch with tag (#7641)
+
+- Support the `DiceCost` used by [K-Net](https://arxiv.org/abs/2106.14855) in `MaskHungarianAssigner` (#7716)
+
+- Support splitting COCO data for Semi-supervised object detection (#7431)
+
+- Support Pathlib for Config.fromfile (#7685)
+
+- Support to use file client in OpenImages dataset (#7433)
+
+- Add a probability parameter to Mosaic transformation (#7371)
+
+- Support specifying interpolation mode in `Resize` pipeline (#7585)
+
+#### Bug Fixes
+
+- Avoid invalid bbox after deform_sampling (#7567)
+- Fix the issue that argument color_theme does not take effect when exporting confusion matrix (#7701)
+- Fix the `end_level` in Necks, which should be the index of the end input backbone level (#7502)
+- Fix the bug that `mix_results` may be None in `MultiImageMixDataset` (#7530)
+- Fix the bug in ResNet plugin when two plugins are used (#7797)
+
+#### Improvements
+
+- Enhance `load_json_logs` of analyze_logs.py for resumed training logs (#7732)
+- Add argument `out_file` in image_demo.py (#7676)
+- Allow mixed precision training with `SimOTAAssigner` (#7516)
+- Updated INF to 100000.0 to be the same as that in the official YOLOX (#7778)
+- Add documentations of:
+  - how to get channels of a new backbone (#7642)
+  - how to unfreeze the backbone network (#7570)
+  - how to train fast_rcnn model (#7549)
+  - proposals in Deformable DETR (#7690)
+  - from-scratch install script in get_started.md (#7575)
+- Release pre-trained models of
+  - [Mask2Former](configs/mask2former) (#7595, #7709)
+  - RetinaNet with ResNet-18 and release models (#7387)
+  - RetinaNet with EfficientNet backbone (#7646)
+
+#### Contributors
+
+A total of 27 developers contributed to this release.
+Thanks @jovialio, @zhangsanfeng2022, @HarryZJ, @jamiechoi1995, @nestiank, @PeterH0323, @RangeKing, @Y-M-Y, @mattcasey02, @weiji14, @Yulv-git, @xiefeifeihu, @FANG-MING, @meng976537406, @nijkah, @sudz123, @CCODING04, @SheffieldCao, @Czm369, @BIGWangYuDong, @zytx121, @jbwang1997, @chhluo, @jshilong, @RangiLyu, @hhaAndroid, @ZwwWayne
+
+### v2.23.0 (28/3/2022)
+
+#### Highlights
+
+- Support Mask2Former: [Masked-attention Mask Transformer for Universal Image Segmentation](https://arxiv.org/abs/2112.01527)
+- Support EfficientNet: [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://arxiv.org/abs/1905.11946)
+- Support setting data root through environment variable `MMDET_DATASETS`, users don't have to modify the corresponding path in config files anymore.
+- Find a good recipe for fine-tuning high precision ResNet backbone pre-trained by Torchvision.
+
+#### New Features
+
+- Support [Mask2Former](configs/mask2former)(#6938)(#7466)(#7471)
+- Support [EfficientNet](configs/efficientnet) (#7514)
+- Support setting data root through environment variable `MMDET_DATASETS`, users don't have to modify the corresponding path in config files anymore. (#7386)
+- Support setting different seeds to different ranks (#7432)
+- Update the `dist_train.sh` so that the script can be used to support launching multi-node training on machines without slurm (#7415)
+- Find a good recipe for fine-tuning high precision ResNet backbone pre-trained by Torchvision (#7489)
+
+#### Bug Fixes
+
+- Fix bug in VOC unit test which removes the data directory (#7270)
+- Adjust the order of `get_classes` and `FileClient` (#7276)
+- Force the inputs of `get_bboxes` in yolox_head to float32 (#7324)
+- Fix misplaced arguments in LoadPanopticAnnotations (#7388)
+- Fix reduction=mean in CELoss. (#7449)
+- Update unit test of CrossEntropyCost (#7537)
+- Fix memory leaking in panpotic segmentation evaluation (#7538)
+- Fix the bug of shape broadcast in YOLOv3 (#7551)
+
+#### Improvements
+
+- Add Chinese version of onnx2tensorrt.md (#7219)
+- Update colab tutorials (#7310)
+- Update information about Localization Distillation (#7350)
+- Add Chinese version of `finetune.md` (#7178)
+- Update YOLOX log for non square input (#7235)
+- Add `nproc` in `coco_panoptic.py` for panoptic quality computing (#7315)
+- Allow to set channel_order in LoadImageFromFile (#7258)
+- Take point sample related functions out of mask_point_head (#7353)
+- Add instance evaluation for coco_panoptic (#7313)
+- Enhance the robustness of analyze_logs.py (#7407)
+- Supplementary notes of sync_random_seed (#7440)
+- Update docstring of cross entropy loss (#7472)
+- Update pascal voc result (#7503)
+- We create How-to documentation to record any questions about How to xxx. In this version, we added
+  - How to use Mosaic augmentation (#7507)
+  - How to use backbone in mmcls (#7438)
+  - How to produce and submit the prediction results of panoptic segmentation models on COCO test-dev set (#7430))
+
+#### Contributors
+
+A total of 27 developers contributed to this release.
+Thanks @ZwwWayne, @haofanwang, @shinya7y, @chhluo, @yangrisheng, @triple-Mu, @jbwang1997, @HikariTJU, @imflash217, @274869388, @zytx121, @matrixgame2018, @jamiechoi1995, @BIGWangYuDong, @JingweiZhang12, @Xiangxu-0103, @hhaAndroid, @jshilong, @osbm, @ceroytres, @bunge-bedstraw-herb, @Youth-Got, @daavoo, @jiangyitong, @RangiLyu, @CCODING04, @yarkable
+
+### v2.22.0 (24/2/2022)
+
+#### Highlights
+
+- Support MaskFormer: [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) (#7212)
+- Support DyHead: [Dynamic Head: Unifying Object Detection Heads with Attentions](https://arxiv.org/abs/2106.08322) (#6823)
+- Release a good recipe of using ResNet in object detectors pre-trained by [ResNet Strikes Back](https://arxiv.org/abs/2110.00476), which consistently brings about 3~4 mAP improvements over RetinaNet, Faster/Mask/Cascade Mask R-CNN (#7001)
+- Support [Open Images Dataset](https://storage.googleapis.com/openimages/web/index.html) (#6331)
+- Support TIMM backbone: [PyTorch Image Models](https://github.com/rwightman/pytorch-image-models) (#7020)
+
+#### New Features
+
+- Support [MaskFormer](configs/maskformer) (#7212)
+- Support [DyHead](configs/dyhead) (#6823)
+- Support [ResNet Strikes Back](configs/resnet_strikes_back) (#7001)
+- Support [OpenImages Dataset](configs/openimages) (#6331)
+- Support [TIMM backbone](configs/timm_example) (#7020)
+- Support visualization for Panoptic Segmentation (#7041)
+
+#### Breaking Changes
+
+In order to support the visualization for Panoptic Segmentation, the `num_classes` can not be `None` when using the `get_palette` function to determine whether to use the panoptic palette.
+
+#### Bug Fixes
+
+- Fix bug for the best checkpoints can not be saved when the `key_score` is None (#7101)
+- Fix MixUp transform filter boxes failing case (#7080)
+- Add missing properties in SABLHead (#7091)
+- Fix bug when NaNs exist in confusion matrix (#7147)
+- Fix PALETTE AttributeError in downstream task (#7230)
+
+#### Improvements
+
+- Speed up SimOTA matching (#7098)
+- Add Chinese translation of `docs_zh-CN/tutorials/init_cfg.md` (#7188)
+
+#### Contributors
+
+A total of 20 developers contributed to this release.
+Thanks @ZwwWayne, @hhaAndroid, @RangiLyu, @AronLin, @BIGWangYuDong, @jbwang1997, @zytx121, @chhluo, @shinya7y, @LuooChen, @dvansa, @siatwangmin, @del-zhenwu, @vikashranjan26, @haofanwang, @jamiechoi1995, @HJoonKwon, @yarkable, @zhijian-liu, @RangeKing
+
+### v2.21.0 (8/2/2022)
+
+### Breaking Changes
+
+To standardize the contents in config READMEs and meta files of OpenMMLab projects, the READMEs and meta files in each config directory have been significantly changed. The template will be released in the future, for now, you can refer to the examples of README for [algorithm](https://github.com/open-mmlab/mmdetection/blob/master/configs/faster_rcnn/README.md), [dataset](https://github.com/open-mmlab/mmdetection/blob/master/configs/deepfashion/README.md) and [backbone](https://github.com/open-mmlab/mmdetection/blob/master/configs/regnet/README.md). To align with the standard, the configs in dcn are put into to two directories named `dcn` and `dcnv2`.
+
+#### New Features
+
+- Allow to customize colors of different classes during visualization (#6716)
+- Support CPU training (#7016)
+- Add download script of COCO, LVIS, and VOC dataset (#7015)
+
+#### Bug Fixes
+
+- Fix weight conversion issue of RetinaNet with Swin-S (#6973)
+- Update `__repr__` of `Compose` (#6951)
+- Fix BadZipFile Error when build docker (#6966)
+- Fix bug in non-distributed multi-gpu training/testing (#7019)
+- Fix bbox clamp in PyTorch 1.10 (#7074)
+- Relax the requirement of PALETTE in dataset wrappers (#7085)
+- Keep the same weights before reassign in the PAA head (#7032)
+- Update code demo in doc (#7092)
+
+#### Improvements
+
+- Speed-up training by allow to set variables of multi-processing (#6974, #7036)
+- Add links of Chinese tutorials in readme (#6897)
+- Disable cv2 multiprocessing by default for acceleration (#6867)
+- Deprecate the support for "python setup.py test" (#6998)
+- Re-organize metafiles and config readmes (#7051)
+- Fix None grad problem during training TOOD by adding `SigmoidGeometricMean` (#7090)
+
+#### Contributors
+
+A total of 26 developers contributed to this release.
+Thanks @del-zhenwu, @zimoqingfeng, @srishilesh, @imyhxy, @jenhaoyang, @jliu-ac, @kimnamu, @ShengliLiu, @garvan2021, @ciusji, @DIYer22, @kimnamu, @q3394101, @zhouzaida, @gaotongxiao, @topsy404, @AntoAndGar, @jbwang1997, @nijkah, @ZwwWayne, @Czm369, @jshilong, @RangiLyu, @BIGWangYuDong, @hhaAndroid, @AronLin
+
+### v2.20.0 (27/12/2021)
+
+#### New Features
+
+- Support [TOOD](configs/tood/README.md): Task-aligned One-stage Object Detection (ICCV 2021 Oral) (#6746)
+- Support resuming from the latest checkpoint automatically (#6727)
+
+#### Bug Fixes
+
+- Fix wrong bbox `loss_weight` of the PAA head (#6744)
+- Fix the padding value of `gt_semantic_seg` in batch collating (#6837)
+- Fix test error of lvis when using `classwise` (#6845)
+- Avoid BC-breaking of `get_local_path`  (#6719)
+- Fix bug in `sync_norm_hook` when the BN layer does not exist (#6852)
+- Use pycocotools directly no matter what platform it is (#6838)
+
+#### Improvements
+
+- Add unit test for SimOTA with no valid bbox (#6770)
+- Use precommit to check readme (#6802)
+- Support selecting GPU-ids in non-distributed testing time (#6781)
+
+#### Contributors
+
+A total of 16 developers contributed to this release.
+Thanks @ZwwWayne, @Czm369, @jshilong, @RangiLyu, @BIGWangYuDong, @hhaAndroid, @jamiechoi1995, @AronLin, @Keiku, @gkagkos, @fcakyon, @www516717402, @vansin, @zactodd, @kimnamu, @jenhaoyang
+
+### v2.19.1 (14/12/2021)
+
+#### New Features
+
+- Release [YOLOX](configs/yolox/README.md) COCO pretrained models (#6698)
+
+#### Bug Fixes
+
+- Fix DCN initialization in DenseHead (#6625)
+- Fix initialization of ConvFCHead (#6624)
+- Fix PseudoSampler in RCNN (#6622)
+- Fix weight initialization in Swin and PVT (#6663)
+- Fix dtype bug in BaseDenseHead (#6767)
+- Fix SimOTA with no valid bbox (#6733)
+
+#### Improvements
+
+- Add an example of combining swin and one-stage models (#6621)
+- Add `get_ann_info` to dataset_wrappers (#6526)
+- Support keeping image ratio in the multi-scale training of YOLOX (#6732)
+- Support `bbox_clip_border` for the augmentations of YOLOX (#6730)
+
+#### Documents
+
+- Update metafile (#6717)
+- Add mmhuman3d in readme (#6699)
+- Update FAQ docs (#6587)
+- Add doc for `detect_anomalous_params` (#6697)
+
+#### Contributors
+
+A total of 11 developers contributed to this release.
+Thanks @ZwwWayne, @LJoson, @Czm369, @jshilong, @ZCMax, @RangiLyu, @BIGWangYuDong, @hhaAndroid, @zhaoxin111, @GT9505, @shinya7y
+
+### v2.19.0 (29/11/2021)
+
+#### Highlights
+
+- Support [Label Assignment Distillation](https://arxiv.org/abs/2108.10520)
+- Support `persistent_workers` for Pytorch >= 1.7
+- Align accuracy to the updated official YOLOX
+
+#### New Features
+
+- Support [Label Assignment Distillation](https://arxiv.org/abs/2108.10520) (#6342)
+- Support `persistent_workers` for Pytorch >= 1.7 (#6435)
+
+#### Bug Fixes
+
+- Fix repeatedly output warning message (#6584)
+- Avoid infinite GPU waiting in dist training (#6501)
+- Fix SSD512 config error (#6574)
+- Fix MMDetection model to ONNX command (#6558)
+
+#### Improvements
+
+- Refactor configs of FP16 models (#6592)
+- Align accuracy to the updated official YOLOX (#6443)
+- Speed up training and reduce memory cost when using PhotoMetricDistortion. (#6442)
+- Make OHEM work with seesaw loss (#6514)
+
+#### Documents
+
+- Update README.md (#6567)
+
+#### Contributors
+
+A total of 11 developers contributed to this release.
+Thanks @FloydHsiu, @RangiLyu, @ZwwWayne, @AndreaPi, @st9007a, @hachreak, @BIGWangYuDong, @hhaAndroid, @AronLin, @chhluo, @vealocia, @HarborYuan, @st9007a, @jshilong
+
+### v2.18.1 (15/11/2021)
+
+#### Highlights
+
+- Release [QueryInst](http://arxiv.org/abs/2105.01928) pre-trained weights (#6460)
+- Support plot confusion matrix (#6344)
+
+#### New Features
+
+- Release [QueryInst](http://arxiv.org/abs/2105.01928) pre-trained weights (#6460)
+- Support plot confusion matrix (#6344)
+
+#### Bug Fixes
+
+- Fix aug test error when the number of prediction bboxes is 0 (#6398)
+- Fix SpatialReductionAttention in PVT (#6488)
+- Fix wrong use of `trunc_normal_init` in PVT and Swin-Transformer (#6432)
+
+#### Improvements
+
+- Save the printed AP information of COCO API to logger (#6505)
+- Always map location to cpu when load checkpoint (#6405)
+- Set a random seed when the user does not set a seed (#6457)
+
+#### Documents
+
+- Chinese version of [Corruption Benchmarking](robustness_benchmarking.md) (#6375)
+- Fix config path in docs (#6396)
+- Update GRoIE readme (#6401)
+
+#### Contributors
+
+A total of 11 developers contributed to this release.
+Thanks @st9007a, @hachreak, @HarborYuan, @vealocia, @chhluo, @AndreaPi, @AronLin, @BIGWangYuDong, @hhaAndroid, @RangiLyu, @ZwwWayne
+
+### v2.18.0 (27/10/2021)
+
+#### Highlights
+
+- Support [QueryInst](http://arxiv.org/abs/2105.01928) (#6050)
+- Refactor dense heads to decouple onnx export logics from `get_bboxes` and speed up inference (#5317, #6003, #6369, #6268, #6315)
+
+#### New Features
+
+- Support [QueryInst](http://arxiv.org/abs/2105.01928) (#6050)
+- Support infinite sampler (#5996)
+
+#### Bug Fixes
+
+- Fix init_weight in fcn_mask_head (#6378)
+- Fix type error in imshow_bboxes of RPN (#6386)
+- Fix broken colab link in MMDetection Tutorial (#6382)
+- Make sure the device and dtype of scale_factor are the same as bboxes (#6374)
+- Remove sampling hardcode (#6317)
+- Fix RandomAffine bbox coordinate recorrection (#6293)
+- Fix init bug of final cls/reg layer in convfc head (#6279)
+- Fix img_shape broken in auto_augment (#6259)
+- Fix kwargs parameter missing error in two_stage (#6256)
+
+#### Improvements
+
+- Unify the interface of stuff head and panoptic head (#6308)
+- Polish readme (#6243)
+- Add code-spell pre-commit hook and fix a typo (#6306)
+- Fix typo (#6245, #6190)
+- Fix sampler unit test (#6284)
+- Fix `forward_dummy` of YOLACT to enable `get_flops` (#6079)
+- Fix link error in the config documentation (#6252)
+- Adjust the order to beautify the document (#6195)
+
+#### Refactors
+
+- Refactor one-stage get_bboxes logic (#5317)
+- Refactor ONNX export of One-Stage models (#6003, #6369)
+- Refactor dense_head and speedup (#6268)
+- Migrate to use prior_generator in training of dense heads (#6315)
+
+#### Contributors
+
+A total of 18 developers contributed to this release.
+Thanks @Boyden, @onnkeat, @st9007a, @vealocia, @yhcao6, @DapangpangX, @yellowdolphin, @cclauss, @kennymckormick,
+@pingguokiller, @collinzrj, @AndreaPi, @AronLin, @BIGWangYuDong, @hhaAndroid, @jshilong, @RangiLyu, @ZwwWayne
+
+### v2.17.0 (28/9/2021)
+
+#### Highlights
+
+- Support [PVT](https://arxiv.org/abs/2102.12122) and [PVTv2](https://arxiv.org/abs/2106.13797)
+- Support [SOLO](https://arxiv.org/abs/1912.04488)
+- Support large scale jittering and New Mask R-CNN baselines
+- Speed up `YOLOv3` inference
+
+#### New Features
+
+- Support [PVT](https://arxiv.org/abs/2102.12122) and [PVTv2](https://arxiv.org/abs/2106.13797) (#5780)
+- Support [SOLO](https://arxiv.org/abs/1912.04488) (#5832)
+- Support large scale jittering and New Mask R-CNN baselines (#6132)
+- Add a general data structure for the results of models (#5508)
+- Added a base class for one-stage instance segmentation (#5904)
+- Speed up `YOLOv3` inference (#5991)
+- Release Swin Transformer pre-trained models (#6100)
+- Support mixed precision training in `YOLOX` (#5983)
+- Support `val` workflow in `YOLACT` (#5986)
+- Add script to test `torchserve` (#5936)
+- Support `onnxsim` with dynamic input shape (#6117)
+
+#### Bug Fixes
+
+- Fix the function naming errors in `model_wrappers` (#5975)
+- Fix regression loss bug when the input is an empty tensor (#5976)
+- Fix scores not contiguous error in `centernet_head` (#6016)
+- Fix missing parameters bug in `imshow_bboxes` (#6034)
+- Fix bug in `aug_test` of `HTC` when the length of `det_bboxes` is 0 (#6088)
+- Fix empty proposal errors in the training of some two-stage models (#5941)
+- Fix `dynamic_axes` parameter error in `ONNX` dynamic shape export (#6104)
+- Fix `dynamic_shape` bug of `SyncRandomSizeHook` (#6144)
+- Fix the Swin Transformer config link error in the configuration (#6172)
+
+#### Improvements
+
+- Add filter rules in `Mosaic` transform (#5897)
+- Add size divisor in get flops to avoid some potential bugs (#6076)
+- Add Chinese translation of `docs_zh-CN/tutorials/customize_dataset.md` (#5915)
+- Add Chinese translation of `conventions.md` (#5825)
+- Add description of the output of data pipeline (#5886)
+- Add dataset information in the README file for `PanopticFPN` (#5996)
+- Add `extra_repr` for `DropBlock` layer to get details in the model printing (#6140)
+- Fix CI out of memory and add PyTorch1.9 Python3.9 unit tests (#5862)
+- Fix download links error of some model (#6069)
+- Improve the generalization of XML dataset (#5943)
+- Polish assertion error messages (#6017)
+- Remove `opencv-python-headless` dependency by `albumentations` (#5868)
+- Check dtype in transform unit tests (#5969)
+- Replace the default theme of documentation with PyTorch Sphinx Theme (#6146)
+- Update the paper and code fields in the metafile (#6043)
+- Support to customize padding value of segmentation map (#6152)
+- Support to resize multiple segmentation maps (#5747)
+
+#### Contributors
+
+A total of 24 developers contributed to this release.
+Thanks @morkovka1337, @HarborYuan, @guillaumefrd, @guigarfr, @www516717402, @gaotongxiao, @ypwhs, @MartaYang, @shinya7y, @justiceeem, @zhaojinjian0000, @VVsssssk, @aravind-anantha, @wangbo-zhao, @czczup, @whai362, @czczup, @marijnl, @AronLin, @BIGWangYuDong, @hhaAndroid, @jshilong, @RangiLyu, @ZwwWayne
+
+### v2.16.0 (30/8/2021)
+
+#### Highlights
+
+- Support [Panoptic FPN](https://arxiv.org/abs/1901.02446) and [Swin Transformer](https://arxiv.org/abs/2103.14030)
+
+#### New Features
+
+- Support [Panoptic FPN](https://arxiv.org/abs/1901.02446) and release models (#5577, #5902)
+- Support Swin Transformer backbone (#5748)
+- Release RetinaNet models pre-trained with multi-scale 3x schedule (#5636)
+- Add script to convert unlabeled image list to coco format (#5643)
+- Add hook to check whether the loss value is valid (#5674)
+- Add YOLO anchor optimizing tool (#5644)
+- Support export onnx models without post process. (#5851)
+- Support classwise evaluation in CocoPanopticDataset (#5896)
+- Adapt browse_dataset for concatenated datasets. (#5935)
+- Add `PatchEmbed` and `PatchMerging` with `AdaptivePadding` (#5952)
+
+#### Bug Fixes
+
+- Fix unit tests of YOLOX (#5859)
+- Fix lose randomness in `imshow_det_bboxes` (#5845)
+- Make output result of `ImageToTensor` contiguous (#5756)
+- Fix inference bug when calling `regress_by_class` in RoIHead in some cases (#5884)
+- Fix bug in CIoU loss where alpha should not have gradient. (#5835)
+- Fix the bug that `multiscale_output` is defined but not used in HRNet (#5887)
+- Set the priority of EvalHook to LOW. (#5882)
+- Fix a YOLOX bug when applying bbox rescaling in test mode (#5899)
+- Fix mosaic coordinate error (#5947)
+- Fix dtype of bbox in RandomAffine. (#5930)
+
+#### Improvements
+
+- Add Chinese version of `data_pipeline` and  (#5662)
+- Support to remove state dicts of EMA when publishing models. (#5858)
+- Refactor the loss function in HTC and SCNet (#5881)
+- Use warnings instead of logger.warning (#5540)
+- Use legacy coordinate in metric of VOC (#5627)
+- Add Chinese version of customize_losses (#5826)
+- Add Chinese version of model_zoo (#5827)
+
+#### Contributors
+
+A total of 19 developers contributed to this release.
+Thanks @ypwhs, @zywvvd, @collinzrj, @OceanPang, @ddonatien, @@haotian-liu, @viibridges, @Muyun99, @guigarfr, @zhaojinjian0000, @jbwang1997,@wangbo-zhao, @xvjiarui, @RangiLyu, @jshilong, @AronLin, @BIGWangYuDong, @hhaAndroid, @ZwwWayne
+
+### v2.15.1 (11/8/2021)
+
+#### Highlights
+
+- Support [YOLOX](https://arxiv.org/abs/2107.08430)
+
+#### New Features
+
+- Support [YOLOX](https://arxiv.org/abs/2107.08430)(#5756, #5758, #5760, #5767, #5770, #5774, #5777, #5808, #5828, #5848)
+
+#### Bug Fixes
+
+- Update correct SSD models. (#5789)
+- Fix casting error in mask structure (#5820)
+- Fix MMCV deployment documentation links. (#5790)
+
+#### Improvements
+
+- Use dynamic MMCV download link in TorchServe dockerfile (#5779)
+- Rename the function `upsample_like` to `interpolate_as` for more general usage (#5788)
+
+#### Contributors
+
+A total of 14 developers contributed to this release.
+Thanks @HAOCHENYE, @xiaohu2015, @HsLOL, @zhiqwang, @Adamdad, @shinya7y, @Johnson-Wang, @RangiLyu, @jshilong, @mmeendez8, @AronLin, @BIGWangYuDong, @hhaAndroid, @ZwwWayne
+
+### v2.15.0 (02/8/2021)
+
+#### Highlights
+
+- Support adding [MIM](https://github.com/open-mmlab/mim) dependencies during pip installation
+- Support MobileNetV2 for SSD-Lite and YOLOv3
+- Support Chinese Documentation
+
+#### New Features
+
+- Add function `upsample_like` (#5732)
+- Support to output pdf and epub format documentation (#5738)
+- Support and release Cascade Mask R-CNN 3x pre-trained models (#5645)
+- Add `ignore_index` to CrossEntropyLoss (#5646)
+- Support adding [MIM](https://github.com/open-mmlab/mim) dependencies during pip installation (#5676)
+- Add MobileNetV2 config and models for YOLOv3 (#5510)
+- Support COCO Panoptic Dataset (#5231)
+- Support ONNX export of cascade models (#5486)
+- Support DropBlock with RetinaNet (#5544)
+- Support MobileNetV2 SSD-Lite (#5526)
+
+#### Bug Fixes
+
+- Fix the device of label in multiclass_nms (#5673)
+- Fix error of backbone initialization from pre-trained checkpoint in config file (#5603, #5550)
+- Fix download links of RegNet pretrained weights (#5655)
+- Fix two-stage runtime error given empty proposal (#5559)
+- Fix flops count error in DETR (#5654)
+- Fix unittest for `NumClassCheckHook` when it is not used. (#5626)
+- Fix description bug of using custom dataset (#5546)
+- Fix bug of `multiclass_nms` that returns the global indices (#5592)
+- Fix `valid_mask` logic error in RPNHead (#5562)
+- Fix unit test error of pretrained configs (#5561)
+- Fix typo error in anchor_head.py (#5555)
+- Fix bug when using dataset wrappers (#5552)
+- Fix a typo error in demo/MMDet_Tutorial.ipynb (#5511)
+- Fixing crash in `get_root_logger` when `cfg.log_level` is not None (#5521)
+- Fix docker version (#5502)
+- Fix optimizer parameter error when using `IterBasedRunner` (#5490)
+
+#### Improvements
+
+- Add unit tests for MMTracking (#5620)
+- Add Chinese translation of documentation (#5718, #5618, #5558, #5423, #5593, #5421, #5408. #5369, #5419, #5530, #5531)
+- Update resource limit (#5697)
+- Update docstring for InstaBoost (#5640)
+- Support key `reduction_override` in all loss functions (#5515)
+- Use repeatdataset to accelerate CenterNet training (#5509)
+- Remove unnecessary code in autoassign (#5519)
+- Add documentation about `init_cfg` (#5273)
+
+#### Contributors
+
+A total of 18 developers contributed to this release.
+Thanks @OceanPang, @AronLin, @hellock, @Outsider565, @RangiLyu, @ElectronicElephant, @likyoo, @BIGWangYuDong, @hhaAndroid, @noobying, @yyz561, @likyoo,
+@zeakey, @ZwwWayne, @ChenyangLiu, @johnson-magic, @qingswu, @BuxianChen
+
+### v2.14.0 (29/6/2021)
+
+#### Highlights
+
+- Add `simple_test` to dense heads to improve the consistency of single-stage and two-stage detectors
+- Revert the `test_mixins` to single image test to improve efficiency and readability
+- Add Faster R-CNN and Mask R-CNN config using multi-scale training with 3x schedule
+
+#### New Features
+
+- Support pretrained models from MoCo v2 and SwAV (#5286)
+- Add Faster R-CNN and Mask R-CNN config using multi-scale training with 3x schedule (#5179, #5233)
+- Add `reduction_override` in MSELoss (#5437)
+- Stable support of exporting DETR to ONNX with dynamic shapes and batch inference (#5168)
+- Stable support of exporting PointRend to ONNX with dynamic shapes and batch inference (#5440)
+
+#### Bug Fixes
+
+- Fix size mismatch bug in `multiclass_nms` (#4980)
+- Fix the import path of `MultiScaleDeformableAttention` (#5338)
+- Fix errors in config of GCNet ResNext101 models (#5360)
+- Fix Grid-RCNN error when there is no bbox result (#5357)
+- Fix errors in `onnx_export` of bbox_head when setting reg_class_agnostic (#5468)
+- Fix type error of AutoAssign in the document (#5478)
+- Fix web links ending with `.md` (#5315)
+
+#### Improvements
+
+- Add `simple_test` to dense heads to improve the consistency of single-stage and two-stage detectors (#5264)
+- Add support for mask diagonal flip in TTA (#5403)
+- Revert the `test_mixins` to single image test to improve efficiency and readability (#5249)
+- Make YOLOv3 Neck more flexible (#5218)
+- Refactor SSD to make it more general (#5291)
+- Refactor `anchor_generator` and `point_generator` (#5349)
+- Allow to configure out the `mask_head` of the HTC algorithm (#5389)
+- Delete deprecated warning in FPN (#5311)
+- Move `model.pretrained` to `model.backbone.init_cfg` (#5370)
+- Make deployment tools more friendly to use (#5280)
+- Clarify installation documentation (#5316)
+- Add ImageNet Pretrained Models docs (#5268)
+- Add FAQ about training loss=nan solution and COCO AP or AR =-1 (# 5312, #5313)
+- Change all weight links of http to https (#5328)
+
+### v2.13.0 (01/6/2021)
+
+#### Highlights
+
+- Support new methods: [CenterNet](https://arxiv.org/abs/1904.07850), [Seesaw Loss](https://arxiv.org/abs/2008.10032), [MobileNetV2](https://arxiv.org/abs/1801.04381)
+
+#### New Features
+
+- Support paper [Objects as Points](https://arxiv.org/abs/1904.07850) (#4602)
+- Support paper [Seesaw Loss for Long-Tailed Instance Segmentation (CVPR 2021)](https://arxiv.org/abs/2008.10032) (#5128)
+- Support [MobileNetV2](https://arxiv.org/abs/1801.04381) backbone and inverted residual block (#5122)
+- Support [MIM](https://github.com/open-mmlab/mim) (#5143)
+- ONNX exportation with dynamic shapes of CornerNet (#5136)
+- Add `mask_soft` config option to allow non-binary masks (#4615)
+- Add PWC metafile (#5135)
+
+#### Bug Fixes
+
+- Fix YOLOv3 FP16 training error (#5172)
+- Fix Cacscade R-CNN TTA test error when `det_bboxes` length is 0  (#5221)
+- Fix `iou_thr` variable naming errors in VOC recall calculation function (#5195)
+- Fix Faster R-CNN performance dropped in ONNX Runtime (#5197)
+- Fix DETR dict changed error when using python 3.8 during iteration  (#5226)
+
+#### Improvements
+
+- Refactor ONNX export of two stage detector (#5205)
+- Replace MMDetection's EvalHook with MMCV's EvalHook for consistency  (#4806)
+- Update RoI extractor for ONNX (#5194)
+- Use better parameter initialization in YOLOv3 head for higher performance (#5181)
+- Release new DCN models of Mask R-CNN by mixed-precision training (#5201)
+- Update YOLOv3 model weights (#5229)
+- Add DetectoRS ResNet-101 model weights (#4960)
+- Discard bboxes with sizes equals to `min_bbox_size` (#5011)
+- Remove duplicated code in DETR head (#5129)
+- Remove unnecessary object in class definition (#5180)
+- Fix doc link (#5192)
+
+### v2.12.0 (01/5/2021)
+
+#### Highlights
+
+- Support new methods: [AutoAssign](https://arxiv.org/abs/2007.03496), [YOLOF](https://arxiv.org/abs/2103.09460), and [Deformable DETR](https://arxiv.org/abs/2010.04159)
+- Stable support of exporting models to ONNX with batched images and dynamic shape (#5039)
+
+#### Backwards Incompatible Changes
+
+MMDetection is going through big refactoring for more general and convenient usages during the releases from v2.12.0 to v2.15.0 (maybe longer).
+In v2.12.0 MMDetection inevitably brings some BC-breakings, including the MMCV dependency, model initialization, model registry, and mask AP evaluation.
+
+- MMCV version. MMDetection v2.12.0 relies on the newest features in MMCV 1.3.3, including `BaseModule` for unified parameter initialization, model registry, and the CUDA operator `MultiScaleDeformableAttn` for [Deformable DETR](https://arxiv.org/abs/2010.04159). Note that MMCV 1.3.2 already contains all the features used by MMDet but has known issues. Therefore, we recommend users skip MMCV v1.3.2 and use v1.3.3, though v1.3.2 might work for most cases.
+- Unified model initialization (#4750). To unify the parameter initialization in OpenMMLab projects, MMCV supports `BaseModule` that accepts `init_cfg` to allow the modules' parameters initialized in a flexible and unified manner. Now the users need to explicitly call `model.init_weights()` in the training script to initialize the model (as in [here](https://github.com/open-mmlab/mmdetection/blob/master/tools/train.py#L162), previously this was handled by the detector. The models in MMDetection have been re-benchmarked to ensure accuracy based on PR #4750. __The downstream projects should update their code accordingly to use MMDetection v2.12.0__.
+- Unified model registry (#5059). To easily use backbones implemented in other OpenMMLab projects, MMDetection migrates to inherit the model registry created in MMCV (#760). In this way, as long as the backbone is supported in an OpenMMLab project and that project also uses the registry in MMCV, users can use that backbone in MMDetection by simply modifying the config without copying the code of that backbone into MMDetection.
+- Mask AP evaluation (#4898). Previous versions calculate the areas of masks through the bounding boxes when calculating the mask AP of small, medium, and large instances. To indeed use the areas of masks, we pop the key `bbox` during mask AP calculation. This change does not affect the overall mask AP evaluation and aligns the mask AP of similar models in other projects like Detectron2.
+
+#### New Features
+
+- Support paper [AutoAssign: Differentiable Label Assignment for Dense Object Detection](https://arxiv.org/abs/2007.03496) (#4295)
+- Support paper [You Only Look One-level Feature](https://arxiv.org/abs/2103.09460) (#4295)
+- Support paper [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159) (#4778)
+- Support calculating IoU with FP16 tensor in `bbox_overlaps` to save memory and keep speed (#4889)
+- Add `__repr__` in custom dataset to count the number of instances (#4756)
+- Add windows support by updating requirements.txt (#5052)
+- Stable support of exporting models to ONNX with batched images and dynamic shape, including SSD, FSAF,FCOS, YOLOv3, RetinaNet, Faster R-CNN, and Mask R-CNN (#5039)
+
+#### Improvements
+
+- Use MMCV `MODEL_REGISTRY` (#5059)
+- Unified parameter initialization for more flexible usage (#4750)
+- Rename variable names and fix docstring in anchor head (#4883)
+- Support training with empty GT in Cascade RPN (#4928)
+- Add more details of usage of `test_robustness` in documentation (#4917)
+- Changing to use `pycocotools` instead of `mmpycocotools` to fully support Detectron2 and MMDetection in one environment (#4939)
+- Update torch serve dockerfile to support dockers of more versions (#4954)
+- Add check for training with single class dataset (#4973)
+- Refactor transformer and DETR Head (#4763)
+- Update FPG model zoo (#5079)
+- More accurate mask AP of small/medium/large instances (#4898)
+
+#### Bug Fixes
+
+- Fix bug in mean_ap.py when calculating mAP by 11 points (#4875)
+- Fix error when key `meta` is not in old checkpoints (#4936)
+- Fix hanging bug when training with empty GT in VFNet, GFL, and FCOS by changing the place of `reduce_mean` (#4923, #4978, #5058)
+- Fix asyncronized inference error and provide related demo (#4941)
+- Fix IoU losses dimensionality unmatch error (#4982)
+- Fix torch.randperm whtn using PyTorch 1.8 (#5014)
+- Fix empty bbox error in `mask_head` when using CARAFE (#5062)
+- Fix `supplement_mask` bug when there are zero-size RoIs (#5065)
+- Fix testing with empty rois in RoI Heads (#5081)
+
+### v2.11.0 (01/4/2021)
+
+__Highlights__
+
+- Support new method: [Localization Distillation for Object Detection](https://arxiv.org/pdf/2102.12252.pdf)
+- Support Pytorch2ONNX with batch inference and dynamic shape
+
+__New Features__
+
+- Support [Localization Distillation for Object Detection](https://arxiv.org/pdf/2102.12252.pdf) (#4758)
+- Support Pytorch2ONNX with batch inference and dynamic shape for Faster-RCNN and mainstream one-stage detectors (#4796)
+
+__Improvements__
+
+- Support batch inference in head of RetinaNet (#4699)
+- Add batch dimension in second stage of Faster-RCNN (#4785)
+- Support batch inference in bbox coder (#4721)
+- Add check for `ann_ids` in `COCODataset` to ensure it is unique (#4789)
+- support for showing the FPN results (#4716)
+- support dynamic shape for grid_anchor (#4684)
+- Move pycocotools version check to when it is used (#4880)
+
+__Bug Fixes__
+
+- Fix a bug of TridentNet when doing the batch inference (#4717)
+- Fix a bug of Pytorch2ONNX in FASF (#4735)
+- Fix a bug when show the image with float type (#4732)
+
+### v2.10.0 (01/03/2021)
+
+#### Highlights
+
+- Support new methods: [FPG](https://arxiv.org/abs/2004.03580)
+- Support ONNX2TensorRT for SSD, FSAF, FCOS, YOLOv3, and Faster R-CNN.
+
+#### New Features
+
+- Support ONNX2TensorRT for SSD, FSAF, FCOS, YOLOv3, and Faster R-CNN (#4569)
+- Support [Feature Pyramid Grids (FPG)](https://arxiv.org/abs/2004.03580) (#4645)
+- Support video demo (#4420)
+- Add seed option for sampler (#4665)
+- Support to customize type of runner (#4570, #4669)
+- Support synchronizing BN buffer in `EvalHook` (#4582)
+- Add script for GIF demo (#4573)
+
+#### Bug Fixes
+
+- Fix ConfigDict AttributeError and add Colab link (#4643)
+- Avoid crash in empty gt training of GFL head (#4631)
+- Fix `iou_thrs` bug in RPN evaluation (#4581)
+- Fix syntax error of config when upgrading model version (#4584)
+
+#### Improvements
+
+- Refactor unit test file structures (#4600)
+- Refactor nms config (#4636)
+- Get loading pipeline by checking the class directly rather than through config strings (#4619)
+- Add doctests for mask target generation and mask structures (#4614)
+- Use deep copy when copying pipeline arguments (#4621)
+- Update documentations (#4642, #4650, #4620, #4630)
+- Remove redundant code calling `import_modules_from_strings` (#4601)
+- Clean deprecated FP16 API (#4571)
+- Check whether `CLASSES` is correctly initialized in the initialization of `XMLDataset` (#4555)
+- Support batch inference in the inference API (#4462, #4526)
+- Clean deprecated warning and fix 'meta' error (#4695)
+
+### v2.9.0 (01/02/2021)
+
+#### Highlights
+
+- Support new methods: [SCNet](https://arxiv.org/abs/2012.10150), [Sparse R-CNN](https://arxiv.org/abs/2011.12450)
+- Move `train_cfg` and `test_cfg` into model in configs
+- Support to visualize results based on prediction quality
+
+#### New Features
+
+- Support [SCNet](https://arxiv.org/abs/2012.10150) (#4356)
+- Support [Sparse R-CNN](https://arxiv.org/abs/2011.12450) (#4219)
+- Support evaluate mAP by multiple IoUs (#4398)
+- Support concatenate dataset for testing (#4452)
+- Support to visualize results based on prediction quality (#4441)
+- Add ONNX simplify option to Pytorch2ONNX script (#4468)
+- Add hook for checking compatibility of class numbers in heads and datasets (#4508)
+
+#### Bug Fixes
+
+- Fix CPU inference bug of Cascade RPN (#4410)
+- Fix NMS error of CornerNet when there is no prediction box (#4409)
+- Fix TypeError in CornerNet inference (#4411)
+- Fix bug of PAA when training with background images (#4391)
+- Fix the error that the window data is not destroyed when `out_file is not None` and `show==False` (#4442)
+- Fix order of NMS `score_factor` that will decrease the performance of YOLOv3 (#4473)
+- Fix bug in HTC TTA when the number of detection boxes is 0 (#4516)
+- Fix resize error in mask data structures (#4520)
+
+#### Improvements
+
+- Allow to customize classes in LVIS dataset (#4382)
+- Add tutorials for building new models with existing datasets (#4396)
+- Add CPU compatibility information in documentation (#4405)
+- Add documentation of deprecated `ImageToTensor` for batch inference (#4408)
+- Add more details in documentation for customizing dataset (#4430)
+- Switch `imshow_det_bboxes` visualization backend from OpenCV to Matplotlib (#4389)
+- Deprecate `ImageToTensor` in `image_demo.py` (#4400)
+- Move train_cfg/test_cfg into model (#4347, #4489)
+- Update docstring for `reg_decoded_bbox` option in bbox heads (#4467)
+- Update dataset information in documentation (#4525)
+- Release pre-trained R50 and R101 PAA detectors with multi-scale 3x training schedules (#4495)
+- Add guidance for speed benchmark (#4537)
+
+### v2.8.0 (04/01/2021)
+
+#### Highlights
+
+- Support new methods: [Cascade RPN](https://arxiv.org/abs/1909.06720), [TridentNet](https://arxiv.org/abs/1901.01892)
+
+#### New Features
+
+- Support [Cascade RPN](https://arxiv.org/abs/1909.06720) (#1900)
+- Support [TridentNet](https://arxiv.org/abs/1901.01892) (#3313)
+
+#### Bug Fixes
+
+- Fix bug of show result in async_benchmark (#4367)
+- Fix scale factor in MaskTestMixin (#4366)
+- Fix but when returning indices in `multiclass_nms` (#4362)
+- Fix bug of empirical attention in resnext backbone error (#4300)
+- Fix bug of `img_norm_cfg` in FCOS-HRNet models with updated performance and models (#4250)
+- Fix invalid checkpoint and log in Mask R-CNN models on Cityscapes dataset (#4287)
+- Fix bug in distributed sampler when dataset is too small (#4257)
+- Fix bug of 'PAFPN has no attribute extra_convs_on_inputs' (#4235)
+
+#### Improvements
+
+- Update model url from aws to aliyun (#4349)
+- Update ATSS for PyTorch 1.6+ (#4359)
+- Update script to install ruby in pre-commit installation (#4360)
+- Delete deprecated `mmdet.ops` (#4325)
+- Refactor hungarian assigner for more general usage in Sparse R-CNN (#4259)
+- Handle scipy import in DETR to reduce package dependencies (#4339)
+- Update documentation of usages for config options after MMCV (1.2.3) supports overriding list in config (#4326)
+- Update pre-train models of faster rcnn trained on COCO subsets (#4307)
+- Avoid zero or too small value for beta in Dynamic R-CNN (#4303)
+- Add doccumentation for Pytorch2ONNX (#4271)
+- Add deprecated warning FPN arguments (#4264)
+- Support returning indices of kept bboxes when using nms (#4251)
+- Update type and device requirements when creating tensors `GFLHead` (#4210)
+- Update device requirements when creating tensors in `CrossEntropyLoss` (#4224)
+
+### v2.7.0 (30/11/2020)
+
+- Support new method: [DETR](https://arxiv.org/abs/2005.12872), [ResNest](https://arxiv.org/abs/2004.08955), Faster R-CNN DC5.
+- Support YOLO, Mask R-CNN, and Cascade R-CNN models exportable to ONNX.
+
+#### New Features
+
+- Support [DETR](https://arxiv.org/abs/2005.12872) (#4201, #4206)
+- Support to link the best checkpoint in training (#3773)
+- Support to override config through options in inference.py (#4175)
+- Support YOLO, Mask R-CNN, and Cascade R-CNN models exportable to ONNX (#4087, #4083)
+- Support [ResNeSt](https://arxiv.org/abs/2004.08955) backbone (#2959)
+- Support unclip border bbox regression (#4076)
+- Add tpfp func in evaluating AP (#4069)
+- Support mixed precision training of SSD detector with other backbones (#4081)
+- Add Faster R-CNN DC5 models (#4043)
+
+#### Bug Fixes
+
+- Fix bug of `gpu_id` in distributed training mode (#4163)
+- Support Albumentations with version higher than 0.5 (#4032)
+- Fix num_classes bug in faster rcnn config (#4088)
+- Update code in docs/2_new_data_model.md (#4041)
+
+#### Improvements
+
+- Ensure DCN offset to have similar type as features in VFNet (#4198)
+- Add config links in README files of models (#4190)
+- Add tutorials for loss conventions (#3818)
+- Add solution to installation issues in 30-series GPUs (#4176)
+- Update docker version in get_started.md (#4145)
+- Add model statistics and polish some titles in configs README (#4140)
+- Clamp neg probability in FreeAnchor (#4082)
+- Speed up expanding large images (#4089)
+- Fix Pytorch 1.7 incompatibility issues (#4103)
+- Update trouble shooting page to resolve segmentation fault (#4055)
+- Update aLRP-Loss in project page (#4078)
+- Clean duplicated `reduce_mean` function (#4056)
+- Refactor Q&A (#4045)
+
+### v2.6.0 (1/11/2020)
+
+- Support new method: [VarifocalNet](https://arxiv.org/abs/2008.13367).
+- Refactored documentation with more tutorials.
+
+#### New Features
+
+- Support GIoU calculation in `BboxOverlaps2D`, and re-implement `giou_loss` using `bbox_overlaps` (#3936)
+- Support random sampling in CPU mode (#3948)
+- Support VarifocalNet (#3666, #4024)
+
+#### Bug Fixes
+
+- Fix SABL validating bug in Cascade R-CNN (#3913)
+- Avoid division by zero in PAA head when num_pos=0 (#3938)
+- Fix temporary directory bug of multi-node testing error (#4034, #4017)
+- Fix `--show-dir` option in test script (#4025)
+- Fix GA-RetinaNet r50 model url (#3983)
+- Update code in docs and fix broken urls (#3947)
+
+#### Improvements
+
+- Refactor pytorch2onnx API into `mmdet.core.export` and use `generate_inputs_and_wrap_model` for pytorch2onnx (#3857, #3912)
+- Update RPN upgrade scripts for v2.5.0 compatibility (#3986)
+- Use mmcv `tensor2imgs` (#4010)
+- Update test robustness (#4000)
+- Update trouble shooting page (#3994)
+- Accelerate PAA training speed (#3985)
+- Support batch_size > 1 in validation (#3966)
+- Use RoIAlign implemented in MMCV for inference in CPU mode (#3930)
+- Documentation refactoring (#4031)
+
+### v2.5.0 (5/10/2020)
+
+#### Highlights
+
+- Support new methods: [YOLACT](https://arxiv.org/abs/1904.02689), [CentripetalNet](https://arxiv.org/abs/2003.09119).
+- Add more documentations for easier and more clear usage.
+
+#### Backwards Incompatible Changes
+
+__FP16 related methods are imported from mmcv instead of mmdet. (#3766, #3822)__
+Mixed precision training utils in `mmdet.core.fp16` are moved to `mmcv.runner`, including `force_fp32`, `auto_fp16`, `wrap_fp16_model`, and `Fp16OptimizerHook`. A deprecation warning will be raised if users attempt to import those methods from `mmdet.core.fp16`, and will be finally removed in V2.10.0.
+
+__\[0, N-1\] represents foreground classes and N indicates background classes for all models. (#3221)__
+Before v2.5.0, the background label for RPN is 0, and N for other heads. Now the behavior is consistent for all models. Thus `self.background_labels` in `dense_heads` is removed and all heads use `self.num_classes` to indicate the class index of background labels.
+This change has no effect on the pre-trained models in the v2.x model zoo, but will affect the training of all models with RPN heads. Two-stage detectors whose RPN head uses softmax will be affected because the order of categories is changed.
+
+**Only call `get_subset_by_classes` when `test_mode=True` and `self.filter_empty_gt=True` (#3695)**
+Function `get_subset_by_classes` in dataset is refactored and only filters out images when `test_mode=True` and `self.filter_empty_gt=True`.
+In the original implementation, `get_subset_by_classes` is not related to the flag `self.filter_empty_gt` and will only be called when the classes is set during initialization no matter `test_mode` is `True` or `False`. This brings ambiguous behavior and potential bugs in many cases. After v2.5.0, if `filter_empty_gt=False`, no matter whether the classes are specified in a dataset, the dataset will use all the images in the annotations. If `filter_empty_gt=True` and `test_mode=True`, no matter whether the classes are specified, the dataset will call \`\`get_subset_by_classes\` to check the images and filter out images containing no GT boxes. Therefore, the users should be responsible for the data filtering/cleaning process for the test dataset.
+
+#### New Features
+
+- Test time augmentation for single stage detectors (#3844, #3638)
+- Support to show the name of experiments during training (#3764)
+- Add `Shear`, `Rotate`, `Translate` Augmentation (#3656, #3619, #3687)
+- Add image-only transformations including `Constrast`, `Equalize`, `Color`, and `Brightness`. (#3643)
+- Support [YOLACT](https://arxiv.org/abs/1904.02689) (#3456)
+- Support [CentripetalNet](https://arxiv.org/abs/2003.09119) (#3390)
+- Support PyTorch 1.6 in docker (#3905)
+
+#### Bug Fixes
+
+- Fix the bug of training ATSS when there is no ground truth boxes (#3702)
+- Fix the bug of using Focal Loss when there is `num_pos` is 0 (#3702)
+- Fix the label index mapping in dataset browser (#3708)
+- Fix Mask R-CNN training stuck problem when their is no positive rois (#3713)
+- Fix the bug of `self.rpn_head.test_cfg` in `RPNTestMixin` by using `self.rpn_head` in rpn head (#3808)
+- Fix deprecated `Conv2d` from mmcv.ops (#3791)
+- Fix device bug in RepPoints (#3836)
+- Fix SABL validating bug (#3849)
+- Use `https://download.openmmlab.com/mmcv/dist/index.html` for installing MMCV (#3840)
+- Fix nonzero in NMS for PyTorch 1.6.0 (#3867)
+- Fix the API change bug of PAA (#3883)
+- Fix typo in bbox_flip (#3886)
+- Fix cv2 import error of ligGL.so.1 in Dockerfile (#3891)
+
+#### Improvements
+
+- Change to use `mmcv.utils.collect_env` for collecting environment information to avoid duplicate codes (#3779)
+- Update checkpoint file names to v2.0 models in documentation (#3795)
+- Update tutorials for changing runtime settings (#3778), modifying loss (#3777)
+- Improve the function of `simple_test_bboxes` in SABL (#3853)
+- Convert mask to bool before using it as img's index for robustness and speedup (#3870)
+- Improve documentation of modules and dataset customization (#3821)
+
+### v2.4.0 (5/9/2020)
+
+__Highlights__
+
+- Fix lots of issues/bugs and reorganize the trouble shooting page
+- Support new methods [SABL](https://arxiv.org/abs/1912.04260), [YOLOv3](https://arxiv.org/abs/1804.02767), and [PAA Assign](https://arxiv.org/abs/2007.08103)
+- Support Batch Inference
+- Start to publish `mmdet` package to PyPI since v2.3.0
+- Switch model zoo to download.openmmlab.com
+
+__Backwards Incompatible Changes__
+
+- Support Batch Inference (#3564, #3686, #3705): Since v2.4.0, MMDetection could inference model with multiple images in a single GPU.
+  This change influences all the test APIs in MMDetection and downstream codebases. To help the users migrate their code, we use `replace_ImageToTensor` (#3686) to convert legacy test data pipelines during dataset initialization.
+- Support RandomFlip with horizontal/vertical/diagonal direction (#3608): Since v2.4.0, MMDetection supports horizontal/vertical/diagonal flip in the data augmentation. This influences bounding box, mask, and image transformations in data augmentation process and the process that will map those data back to the original format.
+- Migrate to use `mmlvis` and `mmpycocotools` for COCO and LVIS dataset (#3727). The APIs are fully compatible with the original `lvis` and `pycocotools`. Users need to uninstall the existing pycocotools and lvis packages in their environment first and install `mmlvis` & `mmpycocotools`.
+
+__Bug Fixes__
+
+- Fix default mean/std for onnx (#3491)
+- Fix coco evaluation and add metric items (#3497)
+- Fix typo for install.md (#3516)
+- Fix atss when sampler per gpu is 1 (#3528)
+- Fix import of fuse_conv_bn (#3529)
+- Fix bug of gaussian_target, update unittest of heatmap (#3543)
+- Fixed VOC2012 evaluate (#3553)
+- Fix scale factor bug of rescale (#3566)
+- Fix with_xxx_attributes in base detector (#3567)
+- Fix boxes scaling when number is 0 (#3575)
+- Fix rfp check when neck config is a list (#3591)
+- Fix import of fuse conv bn in benchmark.py (#3606)
+- Fix webcam demo (#3634)
+- Fix typo and itemize issues in tutorial (#3658)
+- Fix error in distributed training when some levels of FPN are not assigned with bounding boxes (#3670)
+- Fix the width and height orders of stride in valid flag generation (#3685)
+- Fix weight initialization bug in Res2Net DCN (#3714)
+- Fix bug in OHEMSampler (#3677)
+
+__New Features__
+
+- Support Cutout augmentation (#3521)
+- Support evaluation on multiple datasets through ConcatDataset (#3522)
+- Support [PAA assign](https://arxiv.org/abs/2007.08103) #(3547)
+- Support eval metric with pickle results (#3607)
+- Support [YOLOv3](https://arxiv.org/abs/1804.02767) (#3083)
+- Support [SABL](https://arxiv.org/abs/1912.04260) (#3603)
+- Support to publish to Pypi in github-action (#3510)
+- Support custom imports (#3641)
+
+__Improvements__
+
+- Refactor common issues in documentation (#3530)
+- Add pytorch 1.6 to CI config (#3532)
+- Add config to runner meta (#3534)
+- Add eval-option flag for testing (#3537)
+- Add init_eval to evaluation hook (#3550)
+- Add include_bkg in ClassBalancedDataset (#3577)
+- Using config's loading in inference_detector (#3611)
+- Add ATSS ResNet-101 models in model zoo (#3639)
+- Update urls to download.openmmlab.com (#3665)
+- Support non-mask training for CocoDataset (#3711)
+
+### v2.3.0 (5/8/2020)
+
+__Highlights__
+
+- The CUDA/C++ operators have been moved to `mmcv.ops`. For backward compatibility `mmdet.ops` is kept as warppers of `mmcv.ops`.
+- Support new methods [CornerNet](https://arxiv.org/abs/1808.01244), [DIOU](https://arxiv.org/abs/1911.08287)/[CIOU](https://arxiv.org/abs/2005.03572) loss, and new dataset: [LVIS V1](https://arxiv.org/abs/1908.03195)
+- Provide more detailed colab training tutorials and more complete documentation.
+- Support to convert RetinaNet from Pytorch to ONNX.
+
+__Bug Fixes__
+
+- Fix the model initialization bug of DetectoRS (#3187)
+- Fix the bug of module names in NASFCOSHead (#3205)
+- Fix the filename bug in publish_model.py (#3237)
+- Fix the dimensionality bug when `inside_flags.any()` is `False` in dense heads (#3242)
+- Fix the bug of forgetting to pass flip directions in `MultiScaleFlipAug` (#3262)
+- Fixed the bug caused by default value of `stem_channels` (#3333)
+- Fix the bug of model checkpoint loading for CPU inference (#3318, #3316)
+- Fix topk bug when box number is smaller than the expected topk number in ATSSAssigner (#3361)
+- Fix the gt priority bug in center_region_assigner.py (#3208)
+- Fix NaN issue of iou calculation in iou_loss.py (#3394)
+- Fix the bug that `iou_thrs` is not actually used during evaluation in coco.py (#3407)
+- Fix test-time augmentation of RepPoints (#3435)
+- Fix runtimeError caused by incontiguous tensor in Res2Net+DCN (#3412)
+
+__New Features__
+
+- Support [CornerNet](https://arxiv.org/abs/1808.01244) (#3036)
+- Support [DIOU](https://arxiv.org/abs/1911.08287)/[CIOU](https://arxiv.org/abs/2005.03572) loss (#3151)
+- Support [LVIS V1](https://arxiv.org/abs/1908.03195) dataset (#)
+- Support customized hooks in training (#3395)
+- Support fp16 training of generalized focal loss (#3410)
+- Support to convert RetinaNet from Pytorch to ONNX (#3075)
+
+__Improvements__
+
+- Support to process ignore boxes in ATSS assigner (#3082)
+- Allow to crop images without ground truth in `RandomCrop` (#3153)
+- Enable the the `Accuracy` module to set threshold (#3155)
+- Refactoring unit tests (#3206)
+- Unify the training settings of `to_float32` and `norm_cfg` in RegNets configs (#3210)
+- Add colab training tutorials for beginners (#3213, #3273)
+- Move CUDA/C++ operators into `mmcv.ops` and keep `mmdet.ops` as warppers for backward compatibility (#3232)(#3457)
+- Update installation scripts in documentation (#3290) and dockerfile (#3320)
+- Support to set image resize backend (#3392)
+- Remove git hash in version file (#3466)
+- Check mmcv version to force version compatibility (#3460)
+
+### v2.2.0 (1/7/2020)
+
+__Highlights__
+
+- Support new methods: [DetectoRS](https://arxiv.org/abs/2006.02334), [PointRend](https://arxiv.org/abs/1912.08193), [Generalized Focal Loss](https://arxiv.org/abs/2006.04388), [Dynamic R-CNN](https://arxiv.org/abs/2004.06002)
+
+__Bug Fixes__
+
+- Fix FreeAnchor when no gt in image (#3176)
+- Clean up deprecated usage of `register_module()` (#3092, #3161)
+- Fix pretrain bug in NAS FCOS (#3145)
+- Fix `num_classes` in SSD (#3142)
+- Fix FCOS warmup (#3119)
+- Fix `rstrip` in `tools/publish_model.py`
+- Fix `flip_ratio` default value in RandomFLip pipeline (#3106)
+- Fix cityscapes eval with ms_rcnn (#3112)
+- Fix RPN softmax (#3056)
+- Fix filename of LVIS@v0.5 (#2998)
+- Fix nan loss by filtering out-of-frame gt_bboxes in COCO (#2999)
+- Fix bug in FSAF (#3018)
+- Add FocalLoss `num_classes` check (#2964)
+- Fix PISA Loss when there are no gts (#2992)
+- Avoid nan in `iou_calculator` (#2975)
+- Prevent possible bugs in loading and transforms caused by shallow copy (#2967)
+
+__New Features__
+
+- Add DetectoRS (#3064)
+- Support Generalize Focal Loss (#3097)
+- Support PointRend (#2752)
+- Support Dynamic R-CNN (#3040)
+- Add DeepFashion dataset (#2968)
+- Implement FCOS training tricks (#2935)
+- Use BaseDenseHead as base class for anchor-base heads (#2963)
+- Add `with_cp` for BasicBlock (#2891)
+- Add `stem_channels` argument for ResNet (#2954)
+
+__Improvements__
+
+- Add anchor free base head (#2867)
+- Migrate to github action (#3137)
+- Add docstring for datasets, pipelines, core modules and methods (#3130, #3125, #3120)
+- Add VOC benchmark (#3060)
+- Add `concat` mode in GRoI (#3098)
+- Remove cmd arg `autorescale-lr` (#3080)
+- Use `len(data['img_metas'])` to indicate `num_samples` (#3073, #3053)
+- Switch to EpochBasedRunner (#2976)
+
+### v2.1.0 (8/6/2020)
+
+__Highlights__
+
+- Support new backbones: [RegNetX](https://arxiv.org/abs/2003.13678), [Res2Net](https://arxiv.org/abs/1904.01169)
+- Support new methods: [NASFCOS](https://arxiv.org/abs/1906.04423), [PISA](https://arxiv.org/abs/1904.04821), [GRoIE](https://arxiv.org/abs/2004.13665)
+- Support new dataset: [LVIS](https://arxiv.org/abs/1908.03195)
+
+__Bug Fixes__
+
+- Change the CLI argument `--validate` to `--no-validate` to enable validation after training epochs by default. (#2651)
+- Add missing cython to docker file (#2713)
+- Fix bug in nms cpu implementation (#2754)
+- Fix bug when showing mask results (#2763)
+- Fix gcc requirement (#2806)
+- Fix bug in async test (#2820)
+- Fix mask encoding-decoding bugs in test API (#2824)
+- Fix bug in test time augmentation (#2858, #2921, #2944)
+- Fix a typo in comment of apis/train (#2877)
+- Fix the bug of returning None when no gt bboxes are in the original image in `RandomCrop`. Fix the bug that misses to handle `gt_bboxes_ignore`, `gt_label_ignore`, and `gt_masks_ignore` in `RandomCrop`, `MinIoURandomCrop` and `Expand` modules. (#2810)
+- Fix bug of `base_channels` of regnet (#2917)
+- Fix the bug of logger when loading pre-trained weights in base detector (#2936)
+
+__New Features__
+
+- Add IoU models (#2666)
+- Add colab demo for inference
+- Support class agnostic nms (#2553)
+- Add benchmark gathering scripts for development only (#2676)
+- Add mmdet-based project links (#2736, #2767, #2895)
+- Add config dump in training (#2779)
+- Add ClassBalancedDataset (#2721)
+- Add res2net backbone (#2237)
+- Support RegNetX models (#2710)
+- Use `mmcv.FileClient` to support different storage backends (#2712)
+- Add ClassBalancedDataset (#2721)
+- Code Release: Prime Sample Attention in Object Detection (CVPR 2020) (#2626)
+- Implement NASFCOS (#2682)
+- Add class weight in CrossEntropyLoss (#2797)
+- Support LVIS dataset (#2088)
+- Support GRoIE (#2584)
+
+__Improvements__
+
+- Allow different x and y strides in anchor heads. (#2629)
+- Make FSAF loss more robust to no gt (#2680)
+- Compute pure inference time instead (#2657) and update inference speed (#2730)
+- Avoided the possibility that a patch with 0 area is cropped. (#2704)
+- Add warnings when deprecated `imgs_per_gpu` is used. (#2700)
+- Add a mask rcnn example for config (#2645)
+- Update model zoo (#2762, #2866, #2876, #2879, #2831)
+- Add `ori_filename` to img_metas and use it in test show-dir (#2612)
+- Use `img_fields` to handle multiple images during image transform (#2800)
+- Add upsample_cfg support in FPN (#2787)
+- Add `['img']` as default `img_fields` for back compatibility (#2809)
+- Rename the pretrained model from `open-mmlab://resnet50_caffe` and `open-mmlab://resnet50_caffe_bgr` to `open-mmlab://detectron/resnet50_caffe` and `open-mmlab://detectron2/resnet50_caffe`. (#2832)
+- Added sleep(2) in test.py to reduce hanging problem (#2847)
+- Support `c10::half` in CARAFE (#2890)
+- Improve documentations (#2918, #2714)
+- Use optimizer constructor in mmcv and clean the original implementation in `mmdet.core.optimizer` (#2947)
+
+### v2.0.0 (6/5/2020)
+
+In this release, we made lots of major refactoring and modifications.
+
+1. __Faster speed__. We optimize the training and inference speed for common models, achieving up to 30% speedup for training and 25% for inference. Please refer to [model zoo](model_zoo.md#comparison-with-detectron2) for details.
+
+2. __Higher performance__. We change some default hyperparameters with no additional cost, which leads to a gain of performance for most models. Please refer to [compatibility](compatibility.md#training-hyperparameters) for details.
+
+3. __More documentation and tutorials__. We add a bunch of documentation and tutorials to help users get started more smoothly. Read it [here](https://mmdetection.readthedocs.io/en/latest/).
+
+4. __Support PyTorch 1.5__. The support for 1.1 and 1.2 is dropped, and we switch to some new APIs.
+
+5. __Better configuration system__. Inheritance is supported to reduce the redundancy of configs.
+
+6. __Better modular design__. Towards the goal of simplicity and flexibility, we simplify some encapsulation while add more other configurable modules like BBoxCoder, IoUCalculator, OptimizerConstructor, RoIHead. Target computation is also included in heads and the call hierarchy is simpler.
+
+7. Support new methods: [FSAF](https://arxiv.org/abs/1903.00621) and PAFPN (part of [PAFPN](https://arxiv.org/abs/1803.01534)).
+
+__Breaking Changes__
+Models training with MMDetection 1.x are not fully compatible with 2.0, please refer to the [compatibility doc](compatibility.md) for the details and how to migrate to the new version.
+
+__Improvements__
+
+- Unify cuda and cpp API for custom ops. (#2277)
+- New config files with inheritance. (#2216)
+- Encapsulate the second stage into RoI heads. (#1999)
+- Refactor GCNet/EmpericalAttention into plugins. (#2345)
+- Set low quality match as an option in IoU-based bbox assigners. (#2375)
+- Change the codebase's coordinate system. (#2380)
+- Refactor the category order in heads. 0 means the first positive class instead of background now. (#2374)
+- Add bbox sampler and assigner registry. (#2419)
+- Speed up the inference of RPN. (#2420)
+- Add `train_cfg` and `test_cfg` as class members in all anchor heads. (#2422)
+- Merge target computation methods into heads. (#2429)
+- Add bbox coder to support different bbox encoding and losses. (#2480)
+- Unify the API for regression loss. (#2156)
+- Refactor Anchor Generator. (#2474)
+- Make `lr` an optional argument for optimizers. (#2509)
+- Migrate to modules and methods in MMCV. (#2502, #2511, #2569, #2572)
+- Support PyTorch 1.5. (#2524)
+- Drop the support for Python 3.5 and use F-string in the codebase. (#2531)
+
+__Bug Fixes__
+
+- Fix the scale factors for resized images without keep the aspect ratio. (#2039)
+- Check if max_num > 0 before slicing in NMS. (#2486)
+- Fix Deformable RoIPool when there is no instance. (#2490)
+- Fix the default value of assigned labels. (#2536)
+- Fix the evaluation of Cityscapes. (#2578)
+
+__New Features__
+
+- Add deep_stem and avg_down option to ResNet, i.e., support ResNetV1d. (#2252)
+- Add L1 loss. (#2376)
+- Support both polygon and bitmap for instance masks. (#2353, #2540)
+- Support CPU mode for inference. (#2385)
+- Add optimizer constructor for complicated configuration of optimizers. (#2397, #2488)
+- Implement PAFPN. (#2392)
+- Support empty tensor input for some modules. (#2280)
+- Support for custom dataset classes without overriding it. (#2408, #2443)
+- Support to train subsets of coco dataset. (#2340)
+- Add iou_calculator to potentially support more IoU calculation methods. (2405)
+- Support class wise mean AP (was removed in the last version). (#2459)
+- Add option to save the testing result images. (#2414)
+- Support MomentumUpdaterHook. (#2571)
+- Add a demo to inference a single image. (#2605)
+
+### v1.1.0 (24/2/2020)
+
+__Highlights__
+
+- Dataset evaluation is rewritten with a unified api, which is used by both evaluation hooks and test scripts.
+- Support new methods: [CARAFE](https://arxiv.org/abs/1905.02188).
+
+__Breaking Changes__
+
+- The new MMDDP inherits from the official DDP, thus the `__init__` api is changed to be the same as official DDP.
+- The `mask_head` field in HTC config files is modified.
+- The evaluation and testing script is updated.
+- In all transforms, instance masks are stored as a numpy array shaped (n, h, w) instead of a list of (h, w) arrays, where n is the number of instances.
+
+__Bug Fixes__
+
+- Fix IOU assigners when ignore_iof_thr > 0 and there is no pred boxes. (#2135)
+- Fix mAP evaluation when there are no ignored boxes. (#2116)
+- Fix the empty RoI input for Deformable RoI Pooling. (#2099)
+- Fix the dataset settings for multiple workflows. (#2103)
+- Fix the warning related to `torch.uint8` in PyTorch 1.4. (#2105)
+- Fix the inference demo on devices other than gpu:0. (#2098)
+- Fix Dockerfile. (#2097)
+- Fix the bug that `pad_val` is unused in Pad transform. (#2093)
+- Fix the albumentation transform when there is no ground truth bbox. (#2032)
+
+__Improvements__
+
+- Use torch instead of numpy for random sampling. (#2094)
+- Migrate to the new MMDDP implementation in MMCV v0.3. (#2090)
+- Add meta information in logs. (#2086)
+- Rewrite Soft NMS with pytorch extension and remove cython as a dependency. (#2056)
+- Rewrite dataset evaluation. (#2042, #2087, #2114, #2128)
+- Use numpy array for masks in transforms. (#2030)
+
+__New Features__
+
+- Implement "CARAFE: Content-Aware ReAssembly of FEatures". (#1583)
+- Add `worker_init_fn()` in data_loader when seed is set. (#2066, #2111)
+- Add logging utils. (#2035)
+
+### v1.0.0 (30/1/2020)
+
+This release mainly improves the code quality and add more docstrings.
+
+__Highlights__
+
+- Documentation is online now: <https://mmdetection.readthedocs.io>.
+- Support new models: [ATSS](https://arxiv.org/abs/1912.02424).
+- DCN is now available with the api `build_conv_layer` and `ConvModule` like the normal conv layer.
+- A tool to collect environment information is available for trouble shooting.
+
+__Bug Fixes__
+
+- Fix the incompatibility of the latest numpy and pycocotools. (#2024)
+- Fix the case when distributed package is unavailable, e.g., on Windows. (#1985)
+- Fix the dimension issue for `refine_bboxes()`. (#1962)
+- Fix the typo when `seg_prefix` is a list. (#1906)
+- Add segmentation map cropping to RandomCrop. (#1880)
+- Fix the return value of `ga_shape_target_single()`. (#1853)
+- Fix the loaded shape of empty proposals. (#1819)
+- Fix the mask data type when using albumentation. (#1818)
+
+__Improvements__
+
+- Enhance AssignResult and SamplingResult. (#1995)
+- Add ability to overwrite existing module in Registry. (#1982)
+- Reorganize requirements and make albumentations and imagecorruptions optional. (#1969)
+- Check NaN in `SSDHead`. (#1935)
+- Encapsulate the DCN in ResNe(X)t into a ConvModule & Conv_layers. (#1894)
+- Refactoring for mAP evaluation and support multiprocessing and logging. (#1889)
+- Init the root logger before constructing Runner to log more information. (#1865)
+- Split `SegResizeFlipPadRescale` into different existing transforms. (#1852)
+- Move `init_dist()` to MMCV. (#1851)
+- Documentation and docstring improvements. (#1971, #1938, #1869, #1838)
+- Fix the color of the same class for mask visualization. (#1834)
+- Remove the option `keep_all_stages` in HTC and Cascade R-CNN. (#1806)
+
+__New Features__
+
+- Add two test-time options `crop_mask` and `rle_mask_encode` for mask heads. (#2013)
+- Support loading grayscale images as single channel. (#1975)
+- Implement "Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection". (#1872)
+- Add sphinx generated docs. (#1859, #1864)
+- Add GN support for flops computation. (#1850)
+- Collect env info for trouble shooting. (#1812)
+
+### v1.0rc1 (13/12/2019)
+
+The RC1 release mainly focuses on improving the user experience, and fixing bugs.
+
+__Highlights__
+
+- Support new models: [FoveaBox](https://arxiv.org/abs/1904.03797), [RepPoints](https://arxiv.org/abs/1904.11490) and [FreeAnchor](https://arxiv.org/abs/1909.02466).
+- Add a Dockerfile.
+- Add a jupyter notebook demo and a webcam demo.
+- Setup the code style and CI.
+- Add lots of docstrings and unit tests.
+- Fix lots of bugs.
+
+__Breaking Changes__
+
+- There was a bug for computing COCO-style mAP w.r.t different scales (AP_s, AP_m, AP_l), introduced by #621. (#1679)
+
+__Bug Fixes__
+
+- Fix a sampling interval bug in Libra R-CNN. (#1800)
+- Fix the learning rate in SSD300 WIDER FACE. (#1781)
+- Fix the scaling issue when `keep_ratio=False`. (#1730)
+- Fix typos. (#1721, #1492, #1242, #1108, #1107)
+- Fix the shuffle argument in `build_dataloader`. (#1693)
+- Clip the proposal when computing mask targets. (#1688)
+- Fix the "index out of range" bug for samplers in some corner cases. (#1610, #1404)
+- Fix the NMS issue on devices other than GPU:0. (#1603)
+- Fix SSD Head and GHM Loss on CPU. (#1578)
+- Fix the OOM error when there are too many gt bboxes. (#1575)
+- Fix the wrong keyword argument `nms_cfg` in HTC. (#1573)
+- Process masks and semantic segmentation in Expand and MinIoUCrop transforms. (#1550, #1361)
+- Fix a scale bug in the Non Local op. (#1528)
+- Fix a bug in transforms when `gt_bboxes_ignore` is None. (#1498)
+- Fix a bug when `img_prefix` is None. (#1497)
+- Pass the device argument to `grid_anchors` and `valid_flags`. (#1478)
+- Fix the data pipeline for test_robustness. (#1476)
+- Fix the argument type of deformable pooling. (#1390)
+- Fix the coco_eval when there are only two classes. (#1376)
+- Fix a bug in Modulated DeformableConv when deformable_group>1. (#1359)
+- Fix the mask cropping in RandomCrop. (#1333)
+- Fix zero outputs in DeformConv when not running on cuda:0. (#1326)
+- Fix the type issue in Expand. (#1288)
+- Fix the inference API. (#1255)
+- Fix the inplace operation in Expand. (#1249)
+- Fix the from-scratch training config. (#1196)
+- Fix inplace add in RoIExtractor which cause an error in PyTorch 1.2. (#1160)
+- Fix FCOS when input images has no positive sample. (#1136)
+- Fix recursive imports. (#1099)
+
+__Improvements__
+
+- Print the config file and mmdet version in the log. (#1721)
+- Lint the code before compiling in travis CI. (#1715)
+- Add a probability argument for the `Expand` transform. (#1651)
+- Update the PyTorch and CUDA version in the docker file. (#1615)
+- Raise a warning when specifying `--validate` in non-distributed training. (#1624, #1651)
+- Beautify the mAP printing. (#1614)
+- Add pre-commit hook. (#1536)
+- Add the argument `in_channels` to backbones. (#1475)
+- Add lots of docstrings and unit tests, thanks to [@Erotemic](https://github.com/Erotemic). (#1603, #1517, #1506, #1505, #1491, #1479, #1477, #1475, #1474)
+- Add support for multi-node distributed test when there is no shared storage. (#1399)
+- Optimize Dockerfile to reduce the image size. (#1306)
+- Update new results of HRNet. (#1284, #1182)
+- Add an argument `no_norm_on_lateral` in FPN. (#1240)
+- Test the compiling in CI. (#1235)
+- Move docs to a separate folder. (#1233)
+- Add a jupyter notebook demo. (#1158)
+- Support different type of dataset for training. (#1133)
+- Use int64_t instead of long in cuda kernels. (#1131)
+- Support unsquare RoIs for bbox and mask heads. (#1128)
+- Manually add type promotion to make compatible to PyTorch 1.2. (#1114)
+- Allowing validation dataset for computing validation loss. (#1093)
+- Use `.scalar_type()` instead of `.type()` to suppress some warnings. (#1070)
+
+__New Features__
+
+- Add an option `--with_ap` to compute the AP for each class. (#1549)
+- Implement "FreeAnchor: Learning to Match Anchors for Visual Object Detection". (#1391)
+- Support [Albumentations](https://github.com/albumentations-team/albumentations) for augmentations in the data pipeline. (#1354)
+- Implement "FoveaBox: Beyond Anchor-based Object Detector". (#1339)
+- Support horizontal and vertical flipping. (#1273, #1115)
+- Implement "RepPoints: Point Set Representation for Object Detection". (#1265)
+- Add test-time augmentation to HTC and Cascade R-CNN. (#1251)
+- Add a COCO result analysis tool. (#1228)
+- Add Dockerfile. (#1168)
+- Add a webcam demo. (#1155, #1150)
+- Add FLOPs counter. (#1127)
+- Allow arbitrary layer order for ConvModule. (#1078)
+
+### v1.0rc0 (27/07/2019)
+
+- Implement lots of new methods and components (Mixed Precision Training, HTC, Libra R-CNN, Guided Anchoring, Empirical Attention, Mask Scoring R-CNN, Grid R-CNN (Plus), GHM, GCNet, FCOS, HRNet, Weight Standardization, etc.). Thank all collaborators!
+- Support two additional datasets: WIDER FACE and Cityscapes.
+- Refactoring for loss APIs and make it more flexible to adopt different losses and related hyper-parameters.
+- Speed up multi-gpu testing.
+- Integrate all compiling and installing in a single script.
+
+### v0.6.0 (14/04/2019)
+
+- Up to 30% speedup compared to the model zoo.
+- Support both PyTorch stable and nightly version.
+- Replace NMS and SigmoidFocalLoss with Pytorch CUDA extensions.
+
+### v0.6rc0(06/02/2019)
+
+- Migrate to PyTorch 1.0.
+
+### v0.5.7 (06/02/2019)
+
+- Add support for Deformable ConvNet v2. (Many thanks to the authors and [@chengdazhi](https://github.com/chengdazhi))
+- This is the last release based on PyTorch 0.4.1.
+
+### v0.5.6 (17/01/2019)
+
+- Add support for Group Normalization.
+- Unify RPNHead and single stage heads (RetinaHead, SSDHead) with AnchorHead.
+
+### v0.5.5 (22/12/2018)
+
+- Add SSD for COCO and PASCAL VOC.
+- Add ResNeXt backbones and detection models.
+- Refactoring for Samplers/Assigners and add OHEM.
+- Add VOC dataset and evaluation scripts.
+
+### v0.5.4 (27/11/2018)
+
+- Add SingleStageDetector and RetinaNet.
+
+### v0.5.3 (26/11/2018)
+
+- Add Cascade R-CNN and Cascade Mask R-CNN.
+- Add support for Soft-NMS in config files.
+
+### v0.5.2 (21/10/2018)
+
+- Add support for custom datasets.
+- Add a script to convert PASCAL VOC annotations to the expected format.
+
+### v0.5.1 (20/10/2018)
+
+- Add BBoxAssigner and BBoxSampler, the `train_cfg` field in config files are restructured.
+- `ConvFCRoIHead` / `SharedFCRoIHead` are renamed to `ConvFCBBoxHead` / `SharedFCBBoxHead` for consistency.

docs/en/notes/compatibility.md

+# Compatibility of MMDetection 2.x
+
+## MMDetection 2.25.0
+
+In order to support Mask2Former for instance segmentation, the original config files of Mask2Former for panpotic segmentation need to be renamed [PR #7571](https://github.com/open-mmlab/mmdetection/pull/7571).
+
+<table align="center">
+    <thead>
+        <tr align='center'>
+            <td>before v2.25.0</td>
+            <td>after v2.25.0</td>
+        </tr>
+    </thead>
+    <tbody><tr valign='top'>
+    <th>
+
+```
+'mask2former_xxx_coco.py' represents config files for **panoptic segmentation**.
+```
+
+</th>
+    <th>
+
+```
+'mask2former_xxx_coco.py' represents config files for **instance segmentation**.
+'mask2former_xxx_coco-panoptic.py' represents config files for **panoptic segmentation**.
+```
+
+</th></tr>
+  </tbody></table>
+
+## MMDetection 2.21.0
+
+In order to support CPU training, the logic of scatter in batch collating has been changed. We recommend to use
+MMCV v1.4.4 or higher. For more details, please refer to [MMCV PR #1621](https://github.com/open-mmlab/mmcv/pull/1621).
+
+## MMDetection 2.18.1
+
+### MMCV compatibility
+
+In order to fix the wrong weight reference bug in BaseTransformerLayer, the logic in batch first mode of MultiheadAttention has been changed.
+We recommend to use MMCV v1.3.17 or higher. For more details, please refer to [MMCV PR #1418](https://github.com/open-mmlab/mmcv/pull/1418).
+
+## MMDetection 2.18.0
+
+### DIIHead compatibility
+
+In order to support QueryInst, attn_feats is added into the returned tuple of DIIHead.
+
+## MMDetection 2.14.0
+
+### MMCV Version
+
+In order to fix the problem that the priority of EvalHook is too low, all hook priorities have been re-adjusted in 1.3.8, so MMDetection 2.14.0 needs to rely on the latest MMCV 1.3.8 version. For related information, please refer to [#1120](https://github.com/open-mmlab/mmcv/pull/1120), for related issues, please refer to [#5343](https://github.com/open-mmlab/mmdetection/issues/5343).
+
+### SSD compatibility
+
+In v2.14.0, to make SSD more flexible to use, [PR5291](https://github.com/open-mmlab/mmdetection/pull/5291) refactored its backbone, neck and head. The users can use the script `tools/model_converters/upgrade_ssd_version.py` to convert their models.
+
+```bash
+python tools/model_converters/upgrade_ssd_version.py ${OLD_MODEL_PATH} ${NEW_MODEL_PATH}
+```
+
+- OLD_MODEL_PATH: the path to load the old version SSD model.
+- NEW_MODEL_PATH: the path to save the converted model weights.
+
+## MMDetection 2.12.0
+
+MMDetection is going through big refactoring for more general and convenient usages during the releases from v2.12.0 to v2.18.0 (maybe longer).
+In v2.12.0 MMDetection inevitably brings some BC-breakings, including the MMCV dependency, model initialization, model registry, and mask AP evaluation.
+
+### MMCV Version
+
+MMDetection v2.12.0 relies on the newest features in MMCV 1.3.3, including `BaseModule` for unified parameter initialization, model registry, and the CUDA operator `MultiScaleDeformableAttn` for [Deformable DETR](https://arxiv.org/abs/2010.04159). Note that MMCV 1.3.2 already contains all the features used by MMDet but has known issues. Therefore, we recommend users to skip MMCV v1.3.2 and use v1.3.2, though v1.3.2 might work for most of the cases.
+
+### Unified model initialization
+
+To unify the parameter initialization in OpenMMLab projects, MMCV supports `BaseModule` that accepts `init_cfg` to allow the modules' parameters initialized in a flexible and unified manner. Now the users need to explicitly call `model.init_weights()` in the training script to initialize the model (as in [here](https://github.com/open-mmlab/mmdetection/blob/main/tools/train.py#L162), previously this was handled by the detector. **The downstream projects must update their model initialization accordingly to use MMDetection v2.12.0**. Please refer to PR #4750 for details.
+
+### Unified model registry
+
+To easily use backbones implemented in other OpenMMLab projects, MMDetection v2.12.0 inherits the model registry created in MMCV (#760). In this way, as long as the backbone is supported in an OpenMMLab project and that project also uses the registry in MMCV, users can use that backbone in MMDetection by simply modifying the config without copying the code of that backbone into MMDetection. Please refer to PR #5059 for more details.
+
+### Mask AP evaluation
+
+Before [PR 4898](https://github.com/open-mmlab/mmdetection/pull/4898) and V2.12.0, the mask AP of small, medium, and large instances is calculated based on the bounding box area rather than the real mask area. This leads to higher `APs` and `APm` but lower `APl` but will not affect the overall mask AP. [PR 4898](https://github.com/open-mmlab/mmdetection/pull/4898) change it to use mask areas by deleting `bbox` in mask AP calculation.
+The new calculation does not affect the overall mask AP evaluation and is consistent with [Detectron2](https://github.com/facebookresearch/detectron2/).
+
+## Compatibility with MMDetection 1.x
+
+MMDetection 2.0 goes through a big refactoring and addresses many legacy issues. It is not compatible with the 1.x version, i.e., running inference with the same model weights in these two versions will produce different results. Thus, MMDetection 2.0 re-benchmarks all the models and provides their links and logs in the model zoo.
+
+The major differences are in four folds: coordinate system, codebase conventions, training hyperparameters, and modular design.
+
+### Coordinate System
+
+The new coordinate system is consistent with [Detectron2](https://github.com/facebookresearch/detectron2/) and treats the center of the most left-top pixel as (0, 0) rather than the left-top corner of that pixel.
+Accordingly, the system interprets the coordinates in COCO bounding box and segmentation annotations as coordinates in range `[0, width]` or `[0, height]`.
+This modification affects all the computation related to the bbox and pixel selection,
+which is more natural and accurate.
+
+- The height and width of a box with corners (x1, y1) and (x2, y2) in the new coordinate system is computed as `width = x2 - x1` and `height = y2 - y1`.
+  In MMDetection 1.x and previous version, a "+ 1" was added both height and width.
+  This modification are in three folds:
+
+  1. Box transformation and encoding/decoding in regression.
+  2. IoU calculation. This affects the matching process between ground truth and bounding box and the NMS process. The effect to compatibility is very negligible, though.
+  3. The corners of bounding box is in float type and no longer quantized. This should provide more accurate bounding box results. This also makes the bounding box and RoIs not required to have minimum size of 1, whose effect is small, though.
+
+- The anchors are center-aligned to feature grid points and in float type.
+  In MMDetection 1.x and previous version, the anchors are in `int` type and not center-aligned.
+  This affects the anchor generation in RPN and all the anchor-based methods.
+
+- ROIAlign is better aligned with the image coordinate system. The new implementation is adopted from [Detectron2](https://github.com/facebookresearch/detectron2/tree/master/detectron2/layers/csrc/ROIAlign).
+  The RoIs are shifted by half a pixel by default when they are used to cropping RoI features, compared to MMDetection 1.x.
+  The old behavior is still available by setting `aligned=False` instead of `aligned=True`.
+
+- Mask cropping and pasting are more accurate.
+
+  1. We use the new RoIAlign to crop mask targets. In MMDetection 1.x, the bounding box is quantized before it is used to crop mask target, and the crop process is implemented by numpy. In new implementation, the bounding box for crop is not quantized and sent to RoIAlign. This implementation accelerates the training speed by a large margin (~0.1s per iter, ~2 hour when training Mask R50 for 1x schedule) and should be more accurate.
+
+  2. In MMDetection 2.0, the "`paste_mask()`" function is different and should be more accurate than those in previous versions. This change follows the modification in [Detectron2](https://github.com/facebookresearch/detectron2/blob/master/detectron2/structures/masks.py) and can improve mask AP on COCO by ~0.5% absolute.
+
+### Codebase Conventions
+
+- MMDetection 2.0 changes the order of class labels to reduce unused parameters in regression and mask branch more naturally (without +1 and -1).
+  This effect all the classification layers of the model to have a different ordering of class labels. The final layers of regression branch and mask head no longer keep K+1 channels for K categories, and their class orders are consistent with the classification branch.
+
+  - In MMDetection 2.0, label "K" means background, and labels \[0, K-1\] correspond to the K = num_categories object categories.
+
+  - In MMDetection 1.x and previous version, label "0" means background, and labels \[1, K\] correspond to the K categories.
+
+  - **Note**: The class order of softmax RPN is still the same as that in 1.x in versions\<=2.4.0 while sigmoid RPN is not affected. The class orders in all heads are unified since MMDetection v2.5.0.
+
+- Low quality matching in R-CNN is not used. In MMDetection 1.x and previous versions, the `max_iou_assigner` will match low quality boxes for each ground truth box in both RPN and R-CNN training. We observe this sometimes does not assign the most perfect GT box to some bounding boxes,
+  thus MMDetection 2.0 do not allow low quality matching by default in R-CNN training in the new system. This sometimes may slightly improve the box AP (~0.1% absolute).
+
+- Separate scale factors for width and height. In MMDetection 1.x and previous versions, the scale factor is a single float in mode `keep_ratio=True`. This is slightly inaccurate because the scale factors for width and height have slight difference. MMDetection 2.0 adopts separate scale factors for width and height, the improvement on AP ~0.1% absolute.
+
+- Configs name conventions are changed. MMDetection V2.0 adopts the new name convention to maintain the gradually growing model zoo as the following:
+
+  ```shell
+  [model]_(model setting)_[backbone]_[neck]_(norm setting)_(misc)_(gpu x batch)_[schedule]_[dataset].py,
+  ```
+
+  where the (`misc`) includes DCN and GCBlock, etc. More details are illustrated in the [documentation for config](tutorials/config)
+
+- MMDetection V2.0 uses new ResNet Caffe backbones to reduce warnings when loading pre-trained models. Most of the new backbones' weights are the same as the former ones but do not have `conv.bias`, except that they use a different `img_norm_cfg`. Thus, the new backbone will not cause warning of unexpected keys.
+
+### Training Hyperparameters
+
+The change in training hyperparameters does not affect
+model-level compatibility but slightly improves the performance. The major ones are:
+
+- The number of proposals after nms is changed from 2000 to 1000 by setting `nms_post=1000` and `max_num=1000`.
+  This slightly improves both mask AP and bbox AP by ~0.2% absolute.
+
+- The default box regression losses for Mask R-CNN, Faster R-CNN and RetinaNet are changed from smooth L1 Loss to L1 loss. This leads to an overall improvement in box AP (~0.6% absolute). However, using L1-loss for other methods such as Cascade R-CNN and HTC does not improve the performance, so we keep the original settings for these methods.
+
+- The sample num of RoIAlign layer is set to be 0 for simplicity. This leads to slightly improvement on mask AP (~0.2% absolute).
+
+- The default setting does not use gradient clipping anymore during training for faster training speed. This does not degrade performance of the most of models. For some models such as RepPoints we keep using gradient clipping to stabilize the training process and to obtain better performance.
+
+- The default warmup ratio is changed from 1/3 to 0.001 for a more smooth warming up process since the gradient clipping is usually not used. The effect is found negligible during our re-benchmarking, though.
+
+### Upgrade Models from 1.x to 2.0
+
+To convert the models trained by MMDetection V1.x to MMDetection V2.0, the users can use the script `tools/model_converters/upgrade_model_version.py` to convert
+their models. The converted models can be run in MMDetection V2.0 with slightly dropped performance (less than 1% AP absolute).
+Details can be found in `configs/legacy`.
+
+## pycocotools compatibility
+
+`mmpycocotools` is the OpenMMlab's fork of official `pycocotools`, which works for both MMDetection and Detectron2.
+Before [PR 4939](https://github.com/open-mmlab/mmdetection/pull/4939), since `pycocotools` and `mmpycocotool` have the same package name, if users already installed `pycocotools` (installed Detectron2 first under the same environment), then the setup of MMDetection will skip installing `mmpycocotool`. Thus MMDetection fails due to the missing `mmpycocotools`.
+If MMDetection is installed before Detectron2, they could work under the same environment.
+[PR 4939](https://github.com/open-mmlab/mmdetection/pull/4939) deprecates mmpycocotools in favor of official pycocotools.
+Users may install MMDetection and Detectron2 under the same environment after [PR 4939](https://github.com/open-mmlab/mmdetection/pull/4939), no matter what the installation order is.

docs/en/notes/contribution_guide.md

+# Contribution

docs/en/notes/faq.md

+# Frequently Asked Questions
+
+We list some common troubles faced by many users and their corresponding solutions here. Feel free to enrich the list if you find any frequent issues and have ways to help others to solve them. If the contents here do not cover your issue, please create an issue using the [provided templates](https://github.com/open-mmlab/mmdetection/blob/main/.github/ISSUE_TEMPLATE/error-report.md/) and make sure you fill in all required information in the template.
+
+## PyTorch 2.0 Support
+
+The vast majority of algorithms in MMDetection now support PyTorch 2.0 and its `torch.compile` function. Users only need to install MMDetection 3.0.0rc7 or later versions to enjoy this feature. If any unsupported algorithms are found during use, please feel free to give us feedback. We also welcome contributions from the community to benchmark the speed improvement brought by using the `torch.compile` function.
+
+To enable the `torch.compile` function, simply add `--cfg-options compile=True` after `train.py` or `test.py`. For example, to enable `torch.compile` for RTMDet, you can use the following command:
+
+```shell
+# Single GPU
+python tools/train.py configs/rtmdet/rtmdet_s_8xb32-300e_coco.py  --cfg-options compile=True
+
+# Single node multiple GPUs
+./tools/dist_train.sh configs/rtmdet/rtmdet_s_8xb32-300e_coco.py 8 --cfg-options compile=True
+
+# Single node multiple GPUs + AMP
+./tools/dist_train.sh configs/rtmdet/rtmdet_s_8xb32-300e_coco.py 8 --cfg-options compile=True --amp
+```
+
+It is important to note that PyTorch 2.0's support for dynamic shapes is not yet fully developed. In most object detection algorithms, not only are the input shapes dynamic, but the loss calculation and post-processing parts are also dynamic. This can lead to slower training speeds when using the `torch.compile` function. Therefore, if you wish to enable the `torch.compile` function, you should follow these principles:
+
+1. Input images to the network are fixed shape, not multi-scale
+2. set `torch._dynamo.config.cache_size_limit` parameter. TorchDynamo will convert and cache the Python bytecode, and the compiled functions will be stored in the cache. When the next check finds that the function needs to be recompiled, the function will be recompiled and cached. However, if the number of recompilations exceeds the maximum value set (64), the function will no longer be cached or recompiled. As mentioned above, the loss calculation and post-processing parts of the object detection algorithm are also dynamically calculated, and these functions need to be recompiled every time. Therefore, setting the `torch._dynamo.config.cache_size_limit` parameter to a smaller value can effectively reduce the compilation time
+
+In MMDetection, you can set the `torch._dynamo.config.cache_size_limit` parameter through the environment variable `DYNAMO_CACHE_SIZE_LIMIT`. For example, the command is as follows:
+
+```shell
+# Single GPU
+export DYNAMO_CACHE_SIZE_LIMIT = 4
+python tools/train.py configs/rtmdet/rtmdet_s_8xb32-300e_coco.py  --cfg-options compile=True
+
+# Single node multiple GPUs
+export DYNAMO_CACHE_SIZE_LIMIT = 4
+./tools/dist_train.sh configs/rtmdet/rtmdet_s_8xb32-300e_coco.py 8 --cfg-options compile=True
+```
+
+About the common questions about PyTorch 2.0's dynamo, you can refer to [here](https://pytorch.org/docs/stable/dynamo/faq.html)
+
+## Installation
+
+Compatibility issue between MMCV and MMDetection; "ConvWS is already registered in conv layer"; "AssertionError: MMCV==xxx is used but incompatible. Please install mmcv>=xxx, \<=xxx."
+
+Compatible MMDetection, MMEngine, and MMCV versions are shown as below. Please choose the correct version of MMCV to avoid installation issues.
+
+| MMDetection version |      MMCV version       |     MMEngine version     |
+| :-----------------: | :---------------------: | :----------------------: |
+|        main         |  mmcv>=2.0.0, \<2.2.0   | mmengine>=0.7.1, \<1.0.0 |
+|        3.3.0        |  mmcv>=2.0.0, \<2.2.0   | mmengine>=0.7.1, \<1.0.0 |
+|        3.2.0        |  mmcv>=2.0.0, \<2.2.0   | mmengine>=0.7.1, \<1.0.0 |
+|        3.1.0        |  mmcv>=2.0.0, \<2.1.0   | mmengine>=0.7.1, \<1.0.0 |
+|        3.0.0        |  mmcv>=2.0.0, \<2.1.0   | mmengine>=0.7.1, \<1.0.0 |
+|      3.0.0rc6       | mmcv>=2.0.0rc4, \<2.1.0 | mmengine>=0.6.0, \<1.0.0 |
+|      3.0.0rc5       | mmcv>=2.0.0rc1, \<2.1.0 | mmengine>=0.3.0, \<1.0.0 |
+|      3.0.0rc4       | mmcv>=2.0.0rc1, \<2.1.0 | mmengine>=0.3.0, \<1.0.0 |
+|      3.0.0rc3       | mmcv>=2.0.0rc1, \<2.1.0 | mmengine>=0.3.0, \<1.0.0 |
+|      3.0.0rc2       | mmcv>=2.0.0rc1, \<2.1.0 | mmengine>=0.1.0, \<1.0.0 |
+|      3.0.0rc1       | mmcv>=2.0.0rc1, \<2.1.0 | mmengine>=0.1.0, \<1.0.0 |
+|      3.0.0rc0       | mmcv>=2.0.0rc1, \<2.1.0 | mmengine>=0.1.0, \<1.0.0 |
+
+**Note:**
+
+1. If you want to install mmdet-v2.x, the compatible MMDetection and MMCV versions table can be found at [here](https://mmdetection.readthedocs.io/en/stable/faq.html#installation). Please choose the correct version of MMCV to avoid installation issues.
+2. In MMCV-v2.x, `mmcv-full` is rename to `mmcv`, if you want to install `mmcv` without CUDA ops, you can install `mmcv-lite`.
+
+- "No module named 'mmcv.ops'"; "No module named 'mmcv.\_ext'".
+
+  1. Uninstall existing `mmcv-lite` in the environment using `pip uninstall mmcv-lite`.
+  2. Install `mmcv` following the [installation instruction](https://mmcv.readthedocs.io/en/2.x/get_started/installation.html).
+
+- "Microsoft Visual C++ 14.0 or graeter is required" during installation on Windows.
+
+  This error happens when building the 'pycocotools.\_mask' extension of pycocotools and the environment lacks corresponding C++ compilation dependencies. You need to download it at Microsoft officials [visual-cpp-build-tools](https://visualstudio.microsoft.com/zh-hans/visual-cpp-build-tools/),  select the "Use C ++ Desktop Development" option to install the minimum dependencies, and then reinstall pycocotools.
+
+- Using Albumentations
+
+  If you would like to use `albumentations`, we suggest using `pip install -r requirements/albu.txt` or
+  `pip install -U albumentations --no-binary qudida,albumentations`.
+  If you simply use `pip install albumentations>=0.3.2`, it will install `opencv-python-headless` simultaneously (even though you have already installed `opencv-python`).
+  Please refer to the [official documentation](https://albumentations.ai/docs/getting_started/installation/#note-on-opencv-dependencies) for details.
+
+- ModuleNotFoundError is raised when using some algorithms
+
+  Some extra dependencies are required for Instaboost, Panoptic Segmentation, LVIS dataset, etc. Please note the error message and install corresponding packages, e.g.,
+
+  ```shell
+  # for instaboost
+  pip install instaboostfast
+  # for panoptic segmentation
+  pip install git+https://github.com/cocodataset/panopticapi.git
+  # for LVIS dataset
+  pip install git+https://github.com/lvis-dataset/lvis-api.git
+  ```
+
+## Coding
+
+- Do I need to reinstall mmdet after some code modifications
+
+  If you follow the best practice and install mmdet with `pip install -e .`, any local modifications made to the code will take effect without reinstallation.
+
+- How to develop with multiple MMDetection versions
+
+  You can have multiple folders like mmdet-3.0, mmdet-3.1.
+  When you run the train or test script, it will adopt the mmdet package in the current folder.
+
+  To use the default MMDetection installed in the environment rather than the one you are working with, you can remove the following line in those scripts:
+
+  ```shell
+  PYTHONPATH="$(dirname $0)/..":$PYTHONPATH
+  ```
+
+## PyTorch/CUDA Environment
+
+- "RTX 30 series card fails when building MMCV or MMDet"
+
+  1. Temporary work-around: do `MMCV_WITH_OPS=1 MMCV_CUDA_ARGS='-gencode=arch=compute_80,code=sm_80' pip install -e .`.
+     The common issue is `nvcc fatal : Unsupported gpu architecture 'compute_86'`. This means that the compiler should optimize for sm_86, i.e., nvidia 30 series card, but such optimizations have not been supported by CUDA toolkit 11.0.
+     This work-around modifies the compile flag by adding `MMCV_CUDA_ARGS='-gencode=arch=compute_80,code=sm_80'`, which tells `nvcc` to optimize for **sm_80**, i.e., Nvidia A100. Although A100 is different from the 30 series card, they use similar ampere architecture. This may hurt the performance but it works.
+  2. PyTorch developers have updated that the default compiler flags should be fixed by [pytorch/pytorch#47585](https://github.com/pytorch/pytorch/pull/47585). So using PyTorch-nightly may also be able to solve the problem, though we have not tested it yet.
+
+- "invalid device function" or "no kernel image is available for execution".
+
+  1. Check if your cuda runtime version (under `/usr/local/`), `nvcc --version` and `conda list cudatoolkit` version match.
+  2. Run `python mmdet/utils/collect_env.py` to check whether PyTorch, torchvision, and MMCV are built for the correct GPU architecture.
+     You may need to set `TORCH_CUDA_ARCH_LIST` to reinstall MMCV.
+     The GPU arch table could be found [here](https://docs.nvidia.com/cuda/cuda-compiler-driver-nvcc/index.html#gpu-feature-list),
+     i.e. run `TORCH_CUDA_ARCH_LIST=7.0 pip install mmcv` to build MMCV for Volta GPUs.
+     The compatibility issue could happen when using old GPUS, e.g., Tesla K80 (3.7) on colab.
+  3. Check whether the running environment is the same as that when mmcv/mmdet has compiled.
+     For example, you may compile mmcv using CUDA 10.0 but run it on CUDA 9.0 environments.
+
+- "undefined symbol" or "cannot open xxx.so".
+
+  1. If those symbols are CUDA/C++ symbols (e.g., libcudart.so or GLIBCXX), check whether the CUDA/GCC runtimes are the same as those used for compiling mmcv,
+     i.e. run `python mmdet/utils/collect_env.py` to see if `"MMCV Compiler"`/`"MMCV CUDA Compiler"` is the same as `"GCC"`/`"CUDA_HOME"`.
+  2. If those symbols are PyTorch symbols (e.g., symbols containing caffe, aten, and TH), check whether the PyTorch version is the same as that used for compiling mmcv.
+  3. Run `python mmdet/utils/collect_env.py` to check whether PyTorch, torchvision, and MMCV are built by and running on the same environment.
+
+- setuptools.sandbox.UnpickleableException: DistutilsSetupError("each element of 'ext_modules' option must be an Extension instance or 2-tuple")
+
+  1. If you are using miniconda rather than anaconda, check whether Cython is installed as indicated in [#3379](https://github.com/open-mmlab/mmdetection/issues/3379).
+     You need to manually install Cython first and then run command `pip install -r requirements.txt`.
+  2. You may also need to check the compatibility between the `setuptools`, `Cython`, and `PyTorch` in your environment.
+
+- "Segmentation fault".
+
+  1. Check you GCC version and use GCC 5.4. This usually caused by the incompatibility between PyTorch and the environment (e.g., GCC \< 4.9 for PyTorch). We also recommend the users to avoid using GCC 5.5 because many feedbacks report that GCC 5.5 will cause "segmentation fault" and simply changing it to GCC 5.4 could solve the problem.
+
+  2. Check whether PyTorch is correctly installed and could use CUDA op, e.g. type the following command in your terminal.
+
+     ```shell
+     python -c 'import torch; print(torch.cuda.is_available())'
+     ```
+
+     And see whether they could correctly output results.
+
+  3. If Pytorch is correctly installed, check whether MMCV is correctly installed.
+
+     ```shell
+     python -c 'import mmcv; import mmcv.ops'
+     ```
+
+     If MMCV is correctly installed, then there will be no issue of the above two commands.
+
+  4. If MMCV and Pytorch is correctly installed, you man use `ipdb`, `pdb` to set breakpoints or directly add 'print' in mmdetection code and see which part leads the segmentation fault.
+
+## Training
+
+- "Loss goes Nan"
+
+  1. Check if the dataset annotations are valid: zero-size bounding boxes will cause the regression loss to be Nan due to the commonly used transformation for box regression. Some small size (width or height are smaller than 1) boxes will also cause this problem after data augmentation (e.g., instaboost). So check the data and try to filter out those zero-size boxes and skip some risky augmentations on the small-size boxes when you face the problem.
+  2. Reduce the learning rate: the learning rate might be too large due to some reasons, e.g., change of batch size. You can rescale them to the value that could stably train the model.
+  3. Extend the warmup iterations: some models are sensitive to the learning rate at the start of the training. You can extend the warmup iterations, e.g., change the `warmup_iters` from 500 to 1000 or 2000.
+  4. Add gradient clipping: some models requires gradient clipping to stabilize the training process. The default of `grad_clip` is `None`, you can add gradient clippint to avoid gradients that are too large, i.e., set `optim_wrapper=dict(clip_grad=dict(max_norm=35, norm_type=2))` in your config file.
+
+- "GPU out of memory"
+
+  1. There are some scenarios when there are large amount of ground truth boxes, which may cause OOM during target assignment. You can set `gpu_assign_thr=N` in the config of assigner thus the assigner will calculate box overlaps through CPU when there are more than N GT boxes.
+
+  2. Set `with_cp=True` in the backbone. This uses the sublinear strategy in PyTorch to reduce GPU memory cost in the backbone.
+
+  3. Try mixed precision training using following the examples in `config/fp16`. The `loss_scale` might need further tuning for different models.
+
+  4. Try to use `AvoidCUDAOOM` to avoid GPU out of memory. It will first retry after calling `torch.cuda.empty_cache()`. If it still fails, it will then retry by converting the type of inputs to FP16 format. If it still fails, it will try to copy inputs from GPUs to CPUs to continue computing. Try AvoidOOM in you code to make the code continue to run when GPU memory runs out:
+
+     ```python
+     from mmdet.utils import AvoidCUDAOOM
+
+     output = AvoidCUDAOOM.retry_if_cuda_oom(some_function)(input1, input2)
+     ```
+
+     You can also try `AvoidCUDAOOM` as a decorator to make the code continue to run when GPU memory runs out:
+
+     ```python
+     from mmdet.utils import AvoidCUDAOOM
+
+     @AvoidCUDAOOM.retry_if_cuda_oom
+     def function(*args, **kwargs):
+         ...
+         return xxx
+     ```
+
+- "RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one"
+
+  1. This error indicates that your module has parameters that were not used in producing loss. This phenomenon may be caused by running different branches in your code in DDP mode.
+  2. You can set `find_unused_parameters = True` in the config to solve the above problems, but this will slow down the training speed.
+  3. You can set `detect_anomalous_params = True` in the config or `model_wrapper_cfg = dict(type='MMDistributedDataParallel', detect_anomalous_params=True)` (More details please refer to [MMEngine](https://github.com/open-mmlab/mmengine/blob/main/mmengine/model/wrappers/distributed.py#L91)) to get the name of those unused parameters. Note `detect_anomalous_params = True` will slow down the training speed, so it is recommended for debugging only.
+
+- Save the best model
+
+  It can be turned on by configuring `default_hooks = dict(checkpoint=dict(type='CheckpointHook', interval=1, save_best='auto'),`. In the case of the `auto` parameter, the first key in the returned evaluation result will be used as the basis for selecting the best model. You can also directly set the key in the evaluation result to manually set it, for example, `save_best='coco/bbox_mAP'`.
+
+## Evaluation
+
+- COCO Dataset, AP or AR = -1
+  1. According to the definition of COCO dataset, the small and medium areas in an image are less than 1024 (32\*32), 9216 (96\*96), respectively.
+  2. If the corresponding area has no object, the result of AP and AR will set to -1.
+
+## Model
+
+- `style` in ResNet
+
+  The `style` parameter in ResNet allows either `pytorch` or `caffe` style. It indicates the difference in the Bottleneck module. Bottleneck is a stacking structure of `1x1-3x3-1x1` convolutional layers. In the case of `caffe` mode, the convolution layer with `stride=2` is the first `1x1` convolution, while in `pyorch` mode, it is the second `3x3` convolution has `stride=2`. A sample code is as below:
+
+  ```python
+  if self.style == 'pytorch':
+        self.conv1_stride = 1
+        self.conv2_stride = stride
+  else:
+        self.conv1_stride = stride
+        self.conv2_stride = 1
+  ```
+
+- ResNeXt parameter description
+
+  ResNeXt comes from the paper [`Aggregated Residual Transformations for Deep Neural Networks`](https://arxiv.org/abs/1611.05431). It introduces  group and uses “cardinality” to control the number of groups to achieve a balance between accuracy and complexity. It controls the basic width and grouping parameters of the internal Bottleneck module through two hyperparameters `baseWidth` and `cardinality`. An example configuration name in MMDetection is `mask_rcnn_x101_64x4d_fpn_mstrain-poly_3x_coco.py`, where `mask_rcnn` represents the algorithm using Mask R-CNN, `x101` represents the backbone network using ResNeXt-101, and `64x4d` represents that the bottleneck block has 64 group and each group has basic width of 4.
+
+- `norm_eval` in backbone
+
+  Since the detection model is usually large and the input image resolution is high, this will result in a small batch of the detection model, which will make the variance of the statistics calculated by BatchNorm during the training process very large and not as stable as the statistics obtained during the pre-training of the backbone network . Therefore, the `norm_eval=True` mode is generally used in training, and the BatchNorm statistics in the pre-trained backbone network are directly used. The few algorithms that use large batches are the `norm_eval=False` mode, such as NASFPN. For the backbone network without ImageNet pre-training and the batch is relatively small, you can consider using `SyncBN`.

docs/en/notes/projects.md

+# Projects based on MMDetection
+
+There are many projects built upon MMDetection.
+We list some of them as examples of how to extend MMDetection for your own projects.
+As the page might not be completed, please feel free to create a PR to update this page.
+
+## Projects as an extension
+
+Some projects extend the boundary of MMDetection for deployment or other research fields.
+They reveal the potential of what MMDetection can do. We list several of them as below.
+
+- [OTEDetection](https://github.com/opencv/mmdetection): OpenVINO training extensions for object detection.
+- [MMDetection3d](https://github.com/open-mmlab/mmdetection3d): OpenMMLab's next-generation platform for general 3D object detection.
+
+## Projects of papers
+
+There are also projects released with papers.
+Some of the papers are published in top-tier conferences (CVPR, ICCV, and ECCV), the others are also highly influential.
+To make this list also a reference for the community to develop and compare new object detection algorithms, we list them following the time order of top-tier conferences.
+Methods already supported and maintained by MMDetection are not listed.
+
+- Involution: Inverting the Inherence of Convolution for Visual Recognition, CVPR21. [\[paper\]](https://arxiv.org/abs/2103.06255)[\[github\]](https://github.com/d-li14/involution)
+- Multiple Instance Active Learning for Object Detection, CVPR 2021. [\[paper\]](https://openaccess.thecvf.com/content/CVPR2021/papers/Yuan_Multiple_Instance_Active_Learning_for_Object_Detection_CVPR_2021_paper.pdf)[\[github\]](https://github.com/yuantn/MI-AOD)
+- Adaptive Class Suppression Loss for Long-Tail Object Detection, CVPR 2021. [\[paper\]](https://arxiv.org/abs/2104.00885)[\[github\]](https://github.com/CASIA-IVA-Lab/ACSL)
+- Generalizable Pedestrian Detection: The Elephant In The Room, CVPR2021. [\[paper\]](https://arxiv.org/abs/2003.08799)[\[github\]](https://github.com/hasanirtiza/Pedestron)
+- Group Fisher Pruning for Practical Network Compression, ICML2021. [\[paper\]](https://github.com/jshilong/FisherPruning/blob/main/resources/paper.pdf)[\[github\]](https://github.com/jshilong/FisherPruning)
+- Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmax, CVPR2020. [\[paper\]](http://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Overcoming_Classifier_Imbalance_for_Long-Tail_Object_Detection_With_Balanced_Group_CVPR_2020_paper.pdf)[\[github\]](https://github.com/FishYuLi/BalancedGroupSoftmax)
+- Coherent Reconstruction of Multiple Humans from a Single Image, CVPR2020. [\[paper\]](https://jiangwenpl.github.io/multiperson/)[\[github\]](https://github.com/JiangWenPL/multiperson)
+- Look-into-Object: Self-supervised Structure Modeling for Object Recognition, CVPR 2020. [\[paper\]](http://openaccess.thecvf.com/content_CVPR_2020/papers/Zhou_Look-Into-Object_Self-Supervised_Structure_Modeling_for_Object_Recognition_CVPR_2020_paper.pdf)[\[github\]](https://github.com/JDAI-CV/LIO)
+- Video Panoptic Segmentation, CVPR2020. [\[paper\]](https://arxiv.org/abs/2006.11339)[\[github\]](https://github.com/mcahny/vps)
+- D2Det: Towards High Quality Object Detection and Instance Segmentation, CVPR2020. [\[paper\]](http://openaccess.thecvf.com/content_CVPR_2020/html/Cao_D2Det_Towards_High_Quality_Object_Detection_and_Instance_Segmentation_CVPR_2020_paper.html)[\[github\]](https://github.com/JialeCao001/D2Det)
+- CentripetalNet: Pursuing High-quality Keypoint Pairs for Object Detection, CVPR2020. [\[paper\]](https://arxiv.org/abs/2003.09119)[\[github\]](https://github.com/KiveeDong/CentripetalNet)
+- Learning a Unified Sample Weighting Network for Object Detection, CVPR 2020. [\[paper\]](http://openaccess.thecvf.com/content_CVPR_2020/html/Cai_Learning_a_Unified_Sample_Weighting_Network_for_Object_Detection_CVPR_2020_paper.html)[\[github\]](https://github.com/caiqi/sample-weighting-network)
+- Scale-equalizing Pyramid Convolution for Object Detection, CVPR2020. [\[paper\]](https://arxiv.org/abs/2005.03101) [\[github\]](https://github.com/jshilong/SEPC)
+- Revisiting the Sibling Head in Object Detector, CVPR2020. [\[paper\]](https://arxiv.org/abs/2003.07540)[\[github\]](https://github.com/Sense-X/TSD)
+- PolarMask: Single Shot Instance Segmentation with Polar Representation, CVPR2020. [\[paper\]](https://arxiv.org/abs/1909.13226)[\[github\]](https://github.com/xieenze/PolarMask)
+- Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection, CVPR2020. [\[paper\]](https://arxiv.org/abs/2003.11818)[\[github\]](https://github.com/ggjy/HitDet.pytorch)
+- ZeroQ: A Novel Zero Shot Quantization Framework, CVPR2020. [\[paper\]](https://arxiv.org/abs/2001.00281)[\[github\]](https://github.com/amirgholami/ZeroQ)
+- CBNet: A Novel Composite Backbone Network Architecture for Object Detection, AAAI2020. [\[paper\]](https://aaai.org/Papers/AAAI/2020GB/AAAI-LiuY.1833.pdf)[\[github\]](https://github.com/VDIGPKU/CBNet)
+- RDSNet: A New Deep Architecture for Reciprocal Object Detection and Instance Segmentation, AAAI2020. [\[paper\]](https://arxiv.org/abs/1912.05070)[\[github\]](https://github.com/wangsr126/RDSNet)
+- Training-Time-Friendly Network for Real-Time Object Detection, AAAI2020. [\[paper\]](https://arxiv.org/abs/1909.00700)[\[github\]](https://github.com/ZJULearning/ttfnet)
+- Cascade RPN: Delving into High-Quality Region Proposal Network with Adaptive Convolution, NeurIPS 2019. [\[paper\]](https://arxiv.org/abs/1909.06720)[\[github\]](https://github.com/thangvubk/Cascade-RPN)
+- Reasoning R-CNN: Unifying Adaptive Global Reasoning into Large-scale Object Detection, CVPR2019. [\[paper\]](http://openaccess.thecvf.com/content_CVPR_2019/papers/Xu_Reasoning-RCNN_Unifying_Adaptive_Global_Reasoning_Into_Large-Scale_Object_Detection_CVPR_2019_paper.pdf)[\[github\]](https://github.com/chanyn/Reasoning-RCNN)
+- Learning RoI Transformer for Oriented Object Detection in Aerial Images, CVPR2019. [\[paper\]](https://arxiv.org/abs/1812.00155)[\[github\]](https://github.com/dingjiansw101/AerialDetection)
+- SOLO: Segmenting Objects by Locations. [\[paper\]](https://arxiv.org/abs/1912.04488)[\[github\]](https://github.com/WXinlong/SOLO)
+- SOLOv2: Dynamic, Faster and Stronger. [\[paper\]](https://arxiv.org/abs/2003.10152)[\[github\]](https://github.com/WXinlong/SOLO)
+- Dense Peppoints: Representing Visual Objects with Dense Point Sets. [\[paper\]](https://arxiv.org/abs/1912.11473)[\[github\]](https://github.com/justimyhxu/Dense-RepPoints)
+- IterDet: Iterative Scheme for Object Detection in Crowded Environments. [\[paper\]](https://arxiv.org/abs/2005.05708)[\[github\]](https://github.com/saic-vul/iterdet)
+- Cross-Iteration Batch Normalization. [\[paper\]](https://arxiv.org/abs/2002.05712)[\[github\]](https://github.com/Howal/Cross-iterationBatchNorm)
+- A Ranking-based, Balanced Loss Function Unifying Classification and Localisation in Object Detection, NeurIPS2020 [\[paper\]](https://arxiv.org/abs/2009.13592)[\[github\]](https://github.com/kemaloksuz/aLRPLoss)
+- RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder, NeurIPS2020 [\[paper\]](https://arxiv.org/abs/2010.15831)[\[github\]](https://github.com/microsoft/RelationNet2)
+- Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection, CVPR2021[\[paper\]](https://arxiv.org/abs/2011.12885)[\[github\]](https://github.com/implus/GFocalV2)
+- Swin Transformer: Hierarchical Vision Transformer using Shifted Windows, ICCV2021[\[paper\]](https://arxiv.org/abs/2103.14030)[\[github\]](https://github.com/SwinTransformer/)
+- Focal Transformer: Focal Self-attention for Local-Global Interactions in Vision Transformers, NeurIPS2021[\[paper\]](https://arxiv.org/abs/2107.00641)[\[github\]](https://github.com/microsoft/Focal-Transformer)
+- End-to-End Semi-Supervised Object Detection with Soft Teacher, ICCV2021[\[paper\]](https://arxiv.org/abs/2106.09018)[\[github\]](https://github.com/microsoft/SoftTeacher)
+- CBNetV2: A Novel Composite Backbone Network Architecture for Object Detection [\[paper\]](http://arxiv.org/abs/2107.00420)[\[github\]](https://github.com/VDIGPKU/CBNetV2)
+- Instances as Queries, ICCV2021 [\[paper\]](https://openaccess.thecvf.com/content/ICCV2021/papers/Fang_Instances_As_Queries_ICCV_2021_paper.pdf)[\[github\]](https://github.com/hustvl/QueryInst)

docs/en/overview.md

+# OVERVIEW
+
+This chapter introduces you to the framework of MMDetection, and provides links to detailed tutorials about MMDetection.
+
+## What is MMDetection
+
+![image](https://user-images.githubusercontent.com/12907710/137271636-56ba1cd2-b110-4812-8221-b4c120320aa9.png)
+
+MMDetection is an object detection toolbox that contains a rich set of object detection, instance segmentation, and panoptic segmentation methods as well as related components and modules, and below is its whole framework:
+
+MMDetection consists of 7 main parts, apis, structures, datasets, models, engine, evaluation and visualization.
+
+- **apis** provides high-level APIs for model inference.
+- **structures** provides data structures like bbox, mask, and DetDataSample.
+- **datasets** supports various dataset for object detection, instance segmentation, and panoptic segmentation.
+  - **transforms** contains a lot of useful data augmentation transforms.
+  - **samplers** defines different data loader sampling strategy.
+- **models** is the most vital part for detectors and contains different components of a detector.
+  - **detectors** defines all of the detection model classes.
+  - **data_preprocessors** is for preprocessing the input data of the model.
+  - **backbones** contains various backbone networks.
+  - **necks** contains various neck components.
+  - **dense_heads** contains various detection heads that perform dense predictions.
+  - **roi_heads** contains various detection heads that predict from RoIs.
+  - **seg_heads** contains various segmentation heads.
+  - **losses** contains various loss functions.
+  - **task_modules** provides modules for detection tasks. E.g. assigners, samplers, box coders, and prior generators.
+  - **layers** provides some basic neural network layers.
+- **engine** is a part for runtime components.
+  - **runner** provides extensions for [MMEngine's runner](https://mmengine.readthedocs.io/en/latest/tutorials/runner.html).
+  - **schedulers** provides schedulers for adjusting optimization hyperparameters.
+  - **optimizers** provides optimizers and optimizer wrappers.
+  - **hooks** provides various hooks of the runner.
+- **evaluation** provides different metrics for evaluating model performance.
+- **visualization** is for visualizing detection results.
+
+## How to Use this Guide
+
+Here is a detailed step-by-step guide to learn more about MMDetection:
+
+1. For installation instructions, please see [get_started](get_started.md).
+
+2. Refer to the below tutorials for the basic usage of MMDetection.
+
+   - [Train and Test](https://mmdetection.readthedocs.io/en/latest/user_guides/index.html#train-test)
+
+   - [Useful Tools](https://mmdetection.readthedocs.io/en/latest/user_guides/index.html#useful-tools)
+
+3. Refer to the below tutorials to dive deeper:
+
+   - [Basic Concepts](https://mmdetection.readthedocs.io/en/latest/advanced_guides/index.html#basic-concepts)
+   - [Component Customization](https://mmdetection.readthedocs.io/en/latest/advanced_guides/index.html#component-customization)
+
+4. For users of MMDetection 2.x version, we provide a guide to help you adapt to the new version. You can find it in the [migration guide](./migration/migration.md).

docs/en/stat.py

+#!/usr/bin/env python
+import functools as func
+import glob
+import os.path as osp
+import re
+
+import numpy as np
+
+url_prefix = 'https://github.com/open-mmlab/mmdetection/blob/main/configs'
+
+files = sorted(glob.glob('../../configs/*/README.md'))
+
+stats = []
+titles = []
+num_ckpts = 0
+
+for f in files:
+    url = osp.dirname(f.replace('../../configs', url_prefix))
+
+    with open(f, 'r') as content_file:
+        content = content_file.read()
+
+    title = content.split('\n')[0].replace('# ', '').strip()
+    ckpts = set(x.lower().strip()
+                for x in re.findall(r'\[model\]\((https?.*)\)', content))
+
+    if len(ckpts) == 0:
+        continue
+
+    _papertype = [x for x in re.findall(r'\[([A-Z]+)\]', content)]
+    assert len(_papertype) > 0
+    papertype = _papertype[0]
+
+    paper = set([(papertype, title)])
+
+    titles.append(title)
+    num_ckpts += len(ckpts)
+
+    statsmsg = f"""
+\t* [{papertype}] [{title}]({url}) ({len(ckpts)} ckpts)
+"""
+    stats.append((paper, ckpts, statsmsg))
+
+allpapers = func.reduce(lambda a, b: a.union(b), [p for p, _, _ in stats])
+msglist = '\n'.join(x for _, _, x in stats)
+
+papertypes, papercounts = np.unique([t for t, _ in allpapers],
+                                    return_counts=True)
+countstr = '\n'.join(
+    [f'   - {t}: {c}' for t, c in zip(papertypes, papercounts)])
+
+modelzoo = f"""
+# Model Zoo Statistics
+
+* Number of papers: {len(set(titles))}
+{countstr}
+
+* Number of checkpoints: {num_ckpts}
+
+{msglist}
+"""
+
+with open('modelzoo_statistics.md', 'w') as f:
+    f.write(modelzoo)

docs/en/switch_language.md

+## <a href='https://mmdetection.readthedocs.io/en/latest/'>English</a>
+
+## <a href='https://mmdetection.readthedocs.io/zh_CN/latest/'>简体中文</a>

docs/en/user_guides/config.md

+# Learn about Configs
+
+MMDetection and other OpenMMLab repositories use [MMEngine's config system](https://mmengine.readthedocs.io/en/latest/advanced_tutorials/config.html). It has a modular and inheritance design, which is convenient to conduct various experiments.
+
+## Config file content
+
+MMDetection uses a modular design, all modules with different functions can be configured through the config. Taking Mask R-CNN as an example, we will introduce each field in the config according to different function modules:
+
+### Model config
+
+In MMDetection's config, we use `model` to set up detection algorithm components. In addition to neural network components such as `backbone`, `neck`, etc, it also requires `data_preprocessor`, `train_cfg`, and `test_cfg`. `data_preprocessor` is responsible for processing a batch of data output by dataloader. `train_cfg`, and `test_cfg` in the model config are for training and testing hyperparameters of the components.
+
+```python
+model = dict(
+    type='MaskRCNN',  # The name of detector
+    data_preprocessor=dict(  # The config of data preprocessor, usually includes image normalization and padding
+        type='DetDataPreprocessor',  # The type of the data preprocessor, refer to https://mmdetection.readthedocs.io/en/latest/api.html#mmdet.models.data_preprocessors.DetDataPreprocessor
+        mean=[123.675, 116.28, 103.53],  # Mean values used to pre-training the pre-trained backbone models, ordered in R, G, B
+        std=[58.395, 57.12, 57.375],  # Standard variance used to pre-training the pre-trained backbone models, ordered in R, G, B
+        bgr_to_rgb=True,  # whether to convert image from BGR to RGB
+        pad_mask=True,  # whether to pad instance masks
+        pad_size_divisor=32),  # The size of padded image should be divisible by ``pad_size_divisor``
+    backbone=dict(  # The config of backbone
+        type='ResNet',  # The type of backbone network. Refer to https://mmdetection.readthedocs.io/en/latest/api.html#mmdet.models.backbones.ResNet
+        depth=50,  # The depth of backbone, usually it is 50 or 101 for ResNet and ResNext backbones.
+        num_stages=4,  # Number of stages of the backbone.
+        out_indices=(0, 1, 2, 3),  # The index of output feature maps produced in each stage
+        frozen_stages=1,  # The weights in the first stage are frozen
+        norm_cfg=dict(  # The config of normalization layers.
+            type='BN',  # Type of norm layer, usually it is BN or GN
+            requires_grad=True),  # Whether to train the gamma and beta in BN
+        norm_eval=True,  # Whether to freeze the statistics in BN
+        style='pytorch', # The style of backbone, 'pytorch' means that stride 2 layers are in 3x3 Conv, 'caffe' means stride 2 layers are in 1x1 Convs.
+    	init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50')),  # The ImageNet pretrained backbone to be loaded
+    neck=dict(
+        type='FPN',  # The neck of detector is FPN. We also support 'NASFPN', 'PAFPN', etc. Refer to https://mmdetection.readthedocs.io/en/latest/api.html#mmdet.models.necks.FPN for more details.
+        in_channels=[256, 512, 1024, 2048],  # The input channels, this is consistent with the output channels of backbone
+        out_channels=256,  # The output channels of each level of the pyramid feature map
+        num_outs=5),  # The number of output scales
+    rpn_head=dict(
+        type='RPNHead',  # The type of RPN head is 'RPNHead', we also support 'GARPNHead', etc. Refer to https://mmdetection.readthedocs.io/en/latest/api.html#mmdet.models.dense_heads.RPNHead for more details.
+        in_channels=256,  # The input channels of each input feature map, this is consistent with the output channels of neck
+        feat_channels=256,  # Feature channels of convolutional layers in the head.
+        anchor_generator=dict(  # The config of anchor generator
+            type='AnchorGenerator',  # Most of methods use AnchorGenerator, SSD Detectors uses `SSDAnchorGenerator`. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/task_modules/prior_generators/anchor_generator.py#L18 for more details
+            scales=[8],  # Basic scale of the anchor, the area of the anchor in one position of a feature map will be scale * base_sizes
+            ratios=[0.5, 1.0, 2.0],  # The ratio between height and width.
+            strides=[4, 8, 16, 32, 64]),  # The strides of the anchor generator. This is consistent with the FPN feature strides. The strides will be taken as base_sizes if base_sizes is not set.
+        bbox_coder=dict(  # Config of box coder to encode and decode the boxes during training and testing
+            type='DeltaXYWHBBoxCoder',  # Type of box coder. 'DeltaXYWHBBoxCoder' is applied for most of the methods. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/task_modules/coders/delta_xywh_bbox_coder.py#L13 for more details.
+            target_means=[0.0, 0.0, 0.0, 0.0],  # The target means used to encode and decode boxes
+            target_stds=[1.0, 1.0, 1.0, 1.0]),  # The standard variance used to encode and decode boxes
+        loss_cls=dict(  # Config of loss function for the classification branch
+            type='CrossEntropyLoss',  # Type of loss for classification branch, we also support FocalLoss etc. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/losses/cross_entropy_loss.py#L201 for more details
+            use_sigmoid=True,  # RPN usually performs two-class classification, so it usually uses the sigmoid function.
+            loss_weight=1.0),  # Loss weight of the classification branch.
+        loss_bbox=dict(  # Config of loss function for the regression branch.
+            type='L1Loss',  # Type of loss, we also support many IoU Losses and smooth L1-loss, etc. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/losses/smooth_l1_loss.py#L56 for implementation.
+            loss_weight=1.0)),  # Loss weight of the regression branch.
+    roi_head=dict(  # RoIHead encapsulates the second stage of two-stage/cascade detectors.
+        type='StandardRoIHead',
+        bbox_roi_extractor=dict(  # RoI feature extractor for bbox regression.
+            type='SingleRoIExtractor',  # Type of the RoI feature extractor, most of methods uses SingleRoIExtractor. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/roi_heads/roi_extractors/single_level_roi_extractor.py#L13 for details.
+            roi_layer=dict(  # Config of RoI Layer
+                type='RoIAlign',  # Type of RoI Layer, DeformRoIPoolingPack and ModulatedDeformRoIPoolingPack are also supported. Refer to https://mmcv.readthedocs.io/en/latest/api.html#mmcv.ops.RoIAlign for details.
+                output_size=7,  # The output size of feature maps.
+                sampling_ratio=0),  # Sampling ratio when extracting the RoI features. 0 means adaptive ratio.
+            out_channels=256,  # output channels of the extracted feature.
+            featmap_strides=[4, 8, 16, 32]),  # Strides of multi-scale feature maps. It should be consistent with the architecture of the backbone.
+        bbox_head=dict(  # Config of box head in the RoIHead.
+            type='Shared2FCBBoxHead',  # Type of the bbox head, Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/roi_heads/bbox_heads/convfc_bbox_head.py#L220 for implementation details.
+            in_channels=256,  # Input channels for bbox head. This is consistent with the out_channels in roi_extractor
+            fc_out_channels=1024,  # Output feature channels of FC layers.
+            roi_feat_size=7,  # Size of RoI features
+            num_classes=80,  # Number of classes for classification
+            bbox_coder=dict(  # Box coder used in the second stage.
+                type='DeltaXYWHBBoxCoder',  # Type of box coder. 'DeltaXYWHBBoxCoder' is applied for most of the methods.
+                target_means=[0.0, 0.0, 0.0, 0.0],  # Means used to encode and decode box
+                target_stds=[0.1, 0.1, 0.2, 0.2]),  # Standard variance for encoding and decoding. It is smaller since the boxes are more accurate. [0.1, 0.1, 0.2, 0.2] is a conventional setting.
+            reg_class_agnostic=False,  # Whether the regression is class agnostic.
+            loss_cls=dict(  # Config of loss function for the classification branch
+                type='CrossEntropyLoss',  # Type of loss for classification branch, we also support FocalLoss etc.
+                use_sigmoid=False,  # Whether to use sigmoid.
+                loss_weight=1.0),  # Loss weight of the classification branch.
+            loss_bbox=dict(  # Config of loss function for the regression branch.
+                type='L1Loss',  # Type of loss, we also support many IoU Losses and smooth L1-loss, etc.
+                loss_weight=1.0)),  # Loss weight of the regression branch.
+        mask_roi_extractor=dict(  # RoI feature extractor for mask generation.
+            type='SingleRoIExtractor',  # Type of the RoI feature extractor, most of methods uses SingleRoIExtractor.
+            roi_layer=dict(  # Config of RoI Layer that extracts features for instance segmentation
+                type='RoIAlign',  # Type of RoI Layer, DeformRoIPoolingPack and ModulatedDeformRoIPoolingPack are also supported
+                output_size=14,  # The output size of feature maps.
+                sampling_ratio=0),  # Sampling ratio when extracting the RoI features.
+            out_channels=256,  # Output channels of the extracted feature.
+            featmap_strides=[4, 8, 16, 32]),  # Strides of multi-scale feature maps.
+        mask_head=dict(  # Mask prediction head
+            type='FCNMaskHead',  # Type of mask head, refer to https://mmdetection.readthedocs.io/en/latest/api.html#mmdet.models.roi_heads.FCNMaskHead for implementation details.
+            num_convs=4,  # Number of convolutional layers in mask head.
+            in_channels=256,  # Input channels, should be consistent with the output channels of mask roi extractor.
+            conv_out_channels=256,  # Output channels of the convolutional layer.
+            num_classes=80,  # Number of class to be segmented.
+            loss_mask=dict(  # Config of loss function for the mask branch.
+                type='CrossEntropyLoss',  # Type of loss used for segmentation
+                use_mask=True,  # Whether to only train the mask in the correct class.
+                loss_weight=1.0))),  # Loss weight of mask branch.
+    train_cfg = dict(  # Config of training hyperparameters for rpn and rcnn
+        rpn=dict(  # Training config of rpn
+            assigner=dict(  # Config of assigner
+                type='MaxIoUAssigner',  # Type of assigner, MaxIoUAssigner is used for many common detectors. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/task_modules/assigners/max_iou_assigner.py#L14 for more details.
+                pos_iou_thr=0.7,  # IoU >= threshold 0.7 will be taken as positive samples
+                neg_iou_thr=0.3,  # IoU < threshold 0.3 will be taken as negative samples
+                min_pos_iou=0.3,  # The minimal IoU threshold to take boxes as positive samples
+                match_low_quality=True,  # Whether to match the boxes under low quality (see API doc for more details).
+                ignore_iof_thr=-1),  # IoF threshold for ignoring bboxes
+            sampler=dict(  # Config of positive/negative sampler
+                type='RandomSampler',  # Type of sampler, PseudoSampler and other samplers are also supported. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/task_modules/samplers/random_sampler.py#L14 for implementation details.
+                num=256,  # Number of samples
+                pos_fraction=0.5,  # The ratio of positive samples in the total samples.
+                neg_pos_ub=-1,  # The upper bound of negative samples based on the number of positive samples.
+                add_gt_as_proposals=False),  # Whether add GT as proposals after sampling.
+            allowed_border=-1,  # The border allowed after padding for valid anchors.
+            pos_weight=-1,  # The weight of positive samples during training.
+            debug=False),  # Whether to set the debug mode
+        rpn_proposal=dict(  # The config to generate proposals during training
+            nms_across_levels=False,  # Whether to do NMS for boxes across levels. Only work in `GARPNHead`, naive rpn does not support do nms cross levels.
+            nms_pre=2000,  # The number of boxes before NMS
+            nms_post=1000,  # The number of boxes to be kept by NMS. Only work in `GARPNHead`.
+            max_per_img=1000,  # The number of boxes to be kept after NMS.
+            nms=dict( # Config of NMS
+                type='nms',  # Type of NMS
+                iou_threshold=0.7 # NMS threshold
+                ),
+            min_bbox_size=0),  # The allowed minimal box size
+        rcnn=dict(  # The config for the roi heads.
+            assigner=dict(  # Config of assigner for second stage, this is different for that in rpn
+                type='MaxIoUAssigner',  # Type of assigner, MaxIoUAssigner is used for all roi_heads for now. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/task_modules/assigners/max_iou_assigner.py#L14 for more details.
+                pos_iou_thr=0.5,  # IoU >= threshold 0.5 will be taken as positive samples
+                neg_iou_thr=0.5,  # IoU < threshold 0.5 will be taken as negative samples
+                min_pos_iou=0.5,  # The minimal IoU threshold to take boxes as positive samples
+                match_low_quality=False,  # Whether to match the boxes under low quality (see API doc for more details).
+                ignore_iof_thr=-1),  # IoF threshold for ignoring bboxes
+            sampler=dict(
+                type='RandomSampler',  # Type of sampler, PseudoSampler and other samplers are also supported. Refer to https://github.com/open-mmlab/mmdetection/blob/main/mmdet/models/task_modules/samplers/random_sampler.py#L14 for implementation details.
+                num=512,  # Number of samples
+                pos_fraction=0.25,  # The ratio of positive samples in the total samples.
+                neg_pos_ub=-1,  # The upper bound of negative samples based on the number of positive samples.
+                add_gt_as_proposals=True
+            ),  # Whether add GT as proposals after sampling.
+            mask_size=28,  # Size of mask
+            pos_weight=-1,  # The weight of positive samples during training.
+            debug=False)),  # Whether to set the debug mode
+    test_cfg = dict(  # Config for testing hyperparameters for rpn and rcnn
+        rpn=dict(  # The config to generate proposals during testing
+            nms_across_levels=False,  # Whether to do NMS for boxes across levels. Only work in `GARPNHead`, naive rpn does not support do nms cross levels.
+            nms_pre=1000,  # The number of boxes before NMS
+            nms_post=1000,  # The number of boxes to be kept by NMS. Only work in `GARPNHead`.
+            max_per_img=1000,  # The number of boxes to be kept after NMS.
+            nms=dict( # Config of NMS
+                type='nms',  #Type of NMS
+                iou_threshold=0.7 # NMS threshold
+                ),
+            min_bbox_size=0),  # The allowed minimal box size
+        rcnn=dict(  # The config for the roi heads.
+            score_thr=0.05,  # Threshold to filter out boxes
+            nms=dict(  # Config of NMS in the second stage
+                type='nms',  # Type of NMS
+                iou_thr=0.5),  # NMS threshold
+            max_per_img=100,  # Max number of detections of each image
+            mask_thr_binary=0.5)))  # Threshold of mask prediction
+```
+
+### Dataset and evaluator config
+
+[Dataloaders](https://mmengine.readthedocs.io/en/latest/tutorials/dataset.html) are required for the training, validation, and testing of the [runner](https://mmengine.readthedocs.io/en/latest/tutorials/runner.html). Dataset and data pipeline need to be set to build the dataloader. Due to the complexity of this part, we use intermediate variables to simplify the writing of dataloader configs.
+
+```python
+dataset_type = 'CocoDataset'  # Dataset type, this will be used to define the dataset
+data_root = 'data/coco/'  # Root path of data
+backend_args = None # Arguments to instantiate the corresponding file backend
+
+train_pipeline = [  # Training data processing pipeline
+    dict(type='LoadImageFromFile', backend_args=backend_args),  # First pipeline to load images from file path
+    dict(
+        type='LoadAnnotations',  # Second pipeline to load annotations for current image
+        with_bbox=True,  # Whether to use bounding box, True for detection
+        with_mask=True,  # Whether to use instance mask, True for instance segmentation
+        poly2mask=True),  # Whether to convert the polygon mask to instance mask, set False for acceleration and to save memory
+    dict(
+        type='Resize',  # Pipeline that resizes the images and their annotations
+        scale=(1333, 800),  # The largest scale of the images
+        keep_ratio=True  # Whether to keep the ratio between height and width
+        ),
+    dict(
+        type='RandomFlip',  # Augmentation pipeline that flips the images and their annotations
+        prob=0.5),  # The probability to flip
+    dict(type='PackDetInputs')  # Pipeline that formats the annotation data and decides which keys in the data should be packed into data_samples
+]
+test_pipeline = [  # Testing data processing pipeline
+    dict(type='LoadImageFromFile', backend_args=backend_args),  # First pipeline to load images from file path
+    dict(type='Resize', scale=(1333, 800), keep_ratio=True),  # Pipeline that resizes the images
+    dict(
+        type='PackDetInputs',  # Pipeline that formats the annotation data and decides which keys in the data should be packed into data_samples
+        meta_keys=('img_id', 'img_path', 'ori_shape', 'img_shape',
+                   'scale_factor'))
+]
+train_dataloader = dict(   # Train dataloader config
+    batch_size=2,  # Batch size of a single GPU
+    num_workers=2,  # Worker to pre-fetch data for each single GPU
+    persistent_workers=True,  # If ``True``, the dataloader will not shut down the worker processes after an epoch end, which can accelerate training speed.
+    sampler=dict(  # training data sampler
+        type='DefaultSampler',  # DefaultSampler which supports both distributed and non-distributed training. Refer to https://mmengine.readthedocs.io/en/latest/api/generated/mmengine.dataset.DefaultSampler.html#mmengine.dataset.DefaultSampler
+        shuffle=True),  # randomly shuffle the training data in each epoch
+    batch_sampler=dict(type='AspectRatioBatchSampler'),  # Batch sampler for grouping images with similar aspect ratio into a same batch. It can reduce GPU memory cost.
+    dataset=dict(  # Train dataset config
+        type=dataset_type,
+        data_root=data_root,
+        ann_file='annotations/instances_train2017.json',  # Path of annotation file
+        data_prefix=dict(img='train2017/'),  # Prefix of image path
+        filter_cfg=dict(filter_empty_gt=True, min_size=32),  # Config of filtering images and annotations
+        pipeline=train_pipeline,
+        backend_args=backend_args))
+val_dataloader = dict(  # Validation dataloader config
+    batch_size=1,  # Batch size of a single GPU. If batch-size > 1, the extra padding area may influence the performance.
+    num_workers=2,  # Worker to pre-fetch data for each single GPU
+    persistent_workers=True,  # If ``True``, the dataloader will not shut down the worker processes after an epoch end, which can accelerate training speed.
+    drop_last=False,  # Whether to drop the last incomplete batch, if the dataset size is not divisible by the batch size
+    sampler=dict(
+        type='DefaultSampler',
+        shuffle=False),  # not shuffle during validation and testing
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        ann_file='annotations/instances_val2017.json',
+        data_prefix=dict(img='val2017/'),
+        test_mode=True,  # Turn on the test mode of the dataset to avoid filtering annotations or images
+        pipeline=test_pipeline,
+        backend_args=backend_args))
+test_dataloader = val_dataloader  # Testing dataloader config
+```
+
+[Evaluators](https://mmengine.readthedocs.io/en/latest/tutorials/evaluation.html) are used to compute the metrics of the trained model on the validation and testing datasets. The config of evaluators consists of one or a list of metric configs:
+
+```python
+val_evaluator = dict(  # Validation evaluator config
+    type='CocoMetric',  # The coco metric used to evaluate AR, AP, and mAP for detection and instance segmentation
+    ann_file=data_root + 'annotations/instances_val2017.json',  # Annotation file path
+    metric=['bbox', 'segm'],  # Metrics to be evaluated, `bbox` for detection and `segm` for instance segmentation
+    format_only=False,
+    backend_args=backend_args)
+test_evaluator = val_evaluator  # Testing evaluator config
+```
+
+Since the test dataset has no annotation files, the test_dataloader and test_evaluator config in MMDetection are generally equal to the val's. If you want to save the detection results on the test dataset, you can write the config like this:
+
+```python
+# inference on test dataset and
+# format the output results for submission.
+test_dataloader = dict(
+    batch_size=1,
+    num_workers=2,
+    persistent_workers=True,
+    drop_last=False,
+    sampler=dict(type='DefaultSampler', shuffle=False),
+    dataset=dict(
+        type=dataset_type,
+        data_root=data_root,
+        ann_file=data_root + 'annotations/image_info_test-dev2017.json',
+        data_prefix=dict(img='test2017/'),
+        test_mode=True,
+        pipeline=test_pipeline))
+test_evaluator = dict(
+    type='CocoMetric',
+    ann_file=data_root + 'annotations/image_info_test-dev2017.json',
+    metric=['bbox', 'segm'],  # Metrics to be evaluated
+    format_only=True,  # Only format and save the results to coco json file
+    outfile_prefix='./work_dirs/coco_detection/test')  # The prefix of output json files
+```
+
+### Training and testing config
+
+MMEngine's runner uses Loop to control the training, validation, and testing processes.
+Users can set the maximum training epochs and validation intervals with these fields.
+
+```python
+train_cfg = dict(
+    type='EpochBasedTrainLoop',  # The training loop type. Refer to https://github.com/open-mmlab/mmengine/blob/main/mmengine/runner/loops.py
+    max_epochs=12,  # Maximum training epochs
+    val_interval=1)  # Validation intervals. Run validation every epoch.
+val_cfg = dict(type='ValLoop')  # The validation loop type
+test_cfg = dict(type='TestLoop')  # The testing loop type
+```
+
+### Optimization config
+
+`optim_wrapper` is the field to configure optimization-related settings. The optimizer wrapper not only provides the functions of the optimizer, but also supports functions such as gradient clipping, mixed precision training, etc. Find more in [optimizer wrapper tutorial](https://mmengine.readthedocs.io/en/latest/tutorials/optim_wrapper.html).
+
+```python
+optim_wrapper = dict(  # Optimizer wrapper config
+    type='OptimWrapper',  # Optimizer wrapper type, switch to AmpOptimWrapper to enable mixed precision training.
+    optimizer=dict(  # Optimizer config. Support all kinds of optimizers in PyTorch. Refer to https://pytorch.org/docs/stable/optim.html#algorithms
+        type='SGD',  # Stochastic gradient descent optimizer
+        lr=0.02,  # The base learning rate
+        momentum=0.9,  # Stochastic gradient descent with momentum
+        weight_decay=0.0001),  # Weight decay of SGD
+    clip_grad=None,  # Gradient clip option. Set None to disable gradient clip. Find usage in https://mmengine.readthedocs.io/en/latest/tutorials/optimizer.html
+    )
+```
+
+`param_scheduler` is a field that configures methods of adjusting optimization hyperparameters such as learning rate and momentum. Users can combine multiple schedulers to create a desired parameter adjustment strategy. Find more in [parameter scheduler tutorial](https://mmengine.readthedocs.io/en/latest/tutorials/param_scheduler.html) and [parameter scheduler API documents](https://mmengine.readthedocs.io/en/latest/api/generated/mmengine.optim._ParamScheduler.html#mmengine.optim._ParamScheduler)
+
+```python
+param_scheduler = [
+    # Linear learning rate warm-up scheduler
+    dict(
+        type='LinearLR',  # Use linear policy to warmup learning rate
+        start_factor=0.001, # The ratio of the starting learning rate used for warmup
+        by_epoch=False,  # The warmup learning rate is updated by iteration
+        begin=0,  # Start from the first iteration
+        end=500),  # End the warmup at the 500th iteration
+    # The main LRScheduler
+    dict(
+        type='MultiStepLR',  # Use multi-step learning rate policy during training
+        by_epoch=True,  # The learning rate is updated by epoch
+        begin=0,   # Start from the first epoch
+        end=12,  # End at the 12th epoch
+        milestones=[8, 11],  # Epochs to decay the learning rate
+        gamma=0.1)  # The learning rate decay ratio
+]
+```
+
+### Hook config
+
+Users can attach Hooks to training, validation, and testing loops to insert some operations during running. There are two different hook fields, one is `default_hooks` and the other is `custom_hooks`.
+
+`default_hooks` is a dict of hook configs, and they are the hooks must be required at the runtime. They have default priority which should not be modified. If not set, runner will use the default values. To disable a default hook, users can set its config to `None`. Find more in [HOOK](https://mmengine.readthedocs.io/en/latest/tutorials/hook.html).
+
+```python
+default_hooks = dict(
+    timer=dict(type='IterTimerHook'),  # Update the time spent during iteration into message hub
+    logger=dict(type='LoggerHook', interval=50),  # Collect logs from different components of Runner and write them to terminal, JSON file, tensorboard and wandb .etc
+    param_scheduler=dict(type='ParamSchedulerHook'), # update some hyper-parameters of optimizer
+    checkpoint=dict(type='CheckpointHook', interval=1), # Save checkpoints periodically
+    sampler_seed=dict(type='DistSamplerSeedHook'),  # Ensure distributed Sampler shuffle is active
+    visualization=dict(type='DetVisualizationHook'))  # Detection Visualization Hook. Used to visualize validation and testing process prediction results
+```
+
+`custom_hooks` is a list of all other hook configs. Users can develop their own hooks and insert them in this field.
+
+```python
+custom_hooks = []
+```
+
+### Runtime config
+
+```python
+default_scope = 'mmdet'  # The default registry scope to find modules. Refer to https://mmengine.readthedocs.io/en/latest/advanced_tutorials/registry.html
+
+env_cfg = dict(
+    cudnn_benchmark=False,  # Whether to enable cudnn benchmark
+    mp_cfg=dict(  # Multi-processing config
+        mp_start_method='fork',  # Use fork to start multi-processing threads. 'fork' usually faster than 'spawn' but maybe unsafe. See discussion in https://github.com/pytorch/pytorch/issues/1355
+        opencv_num_threads=0),  # Disable opencv multi-threads to avoid system being overloaded
+    dist_cfg=dict(backend='nccl'),  # Distribution configs
+)
+
+vis_backends = [dict(type='LocalVisBackend')]  # Visualization backends. Refer to https://mmengine.readthedocs.io/en/latest/advanced_tutorials/visualization.html
+visualizer = dict(
+    type='DetLocalVisualizer', vis_backends=vis_backends, name='visualizer')
+log_processor = dict(
+    type='LogProcessor',  # Log processor to process runtime logs
+    window_size=50,  # Smooth interval of log values
+    by_epoch=True)  # Whether to format logs with epoch type. Should be consistent with the train loop's type.
+
+log_level = 'INFO'  # The level of logging.
+load_from = None  # Load model checkpoint as a pre-trained model from a given path. This will not resume training.
+resume = False  # Whether to resume from the checkpoint defined in `load_from`. If `load_from` is None, it will resume the latest checkpoint in the `work_dir`.
+```
+
+## Iter-based config
+
+MMEngine's Runner also provides an iter-based training loop except for epoch-based.
+To use iter-based training, users should modify the `train_cfg`, `param_scheduler`, `train_dataloader`, `default_hooks`, and `log_processor`.
+Here is an example of changing an epoch-based RetinaNet config to iter-based: `configs/retinanet/retinanet_r50_fpn_90k_coco.py`
+
+```python
+# Iter-based training config
+train_cfg = dict(
+    _delete_=True,  # Ignore the base config setting (optional)
+    type='IterBasedTrainLoop',  # Use iter-based training loop
+    max_iters=90000,  # Maximum iterations
+    val_interval=10000)  # Validation interval
+
+
+# Change the scheduler to iter-based
+param_scheduler = [
+    dict(
+        type='LinearLR', start_factor=0.001, by_epoch=False, begin=0, end=500),
+    dict(
+        type='MultiStepLR',
+        begin=0,
+        end=90000,
+        by_epoch=False,
+        milestones=[60000, 80000],
+        gamma=0.1)
+]
+
+# Switch to InfiniteSampler to avoid dataloader restart
+train_dataloader = dict(sampler=dict(type='InfiniteSampler'))
+
+# Change the checkpoint saving interval to iter-based
+default_hooks = dict(checkpoint=dict(by_epoch=False, interval=10000))
+
+# Change the log format to iter-based
+log_processor = dict(by_epoch=False)
+```
+
+## Config file inheritance
+
+There are 4 basic component types under `config/_base_`, dataset, model, schedule, default_runtime.
+Many methods could be easily constructed with one of these models like Faster R-CNN, Mask R-CNN, Cascade R-CNN, RPN, SSD.
+The configs that are composed by components from `_base_` are called the _primitive_.
+
+For all configs under the same folder, it is recommended to have only **one** _primitive_ config. All other configs should inherit from the _primitive_ config. In this way, the maximum of inheritance level is 3.
+
+For easy understanding, we recommend contributors to inherit from existing methods.
+For example, if some modification is made based on Faster R-CNN, users may first inherit the basic Faster R-CNN structure by specifying `_base_ = ../faster_rcnn/faster-rcnn_r50_fpn_1x_coco.py`, then modify the necessary fields in the config files.
+
+If you are building an entirely new method that does not share the structure with any of the existing methods, you may create a folder `xxx_rcnn` under `configs`,
+
+Please refer to [mmengine config tutorial](https://mmengine.readthedocs.io/en/latest/advanced_tutorials/config.html) for detailed documentation.
+
+By setting the `_base_` field, we can set which files the current configuration file inherits from.
+
+When `_base_` is a string of a file path, it means inheriting the contents from one config file.
+
+```python
+_base_ = './mask-rcnn_r50_fpn_1x_coco.py'
+```
+
+When `_base_` is a list of multiple file paths, it means inheriting from multiple files.
+
+```python
+_base_ = [
+    '../_base_/models/mask-rcnn_r50_fpn.py',
+    '../_base_/datasets/coco_instance.py',
+    '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py'
+]
+```
+
+If you wish to inspect the config file, you may run `python tools/misc/print_config.py /PATH/TO/CONFIG` to see the complete config.
+
+### Ignore some fields in the base configs
+
+Sometimes, you may set `_delete_=True` to ignore some of the fields in base configs.
+You may refer to [mmengine config tutorial](https://mmengine.readthedocs.io/en/latest/advanced_tutorials/config.html) for a simple illustration.
+
+In MMDetection, for example, to change the backbone of Mask R-CNN with the following config.
+
+```python
+model = dict(
+    type='MaskRCNN',
+    backbone=dict(
+        type='ResNet',
+        depth=50,
+        num_stages=4,
+        out_indices=(0, 1, 2, 3),
+        frozen_stages=1,
+        norm_cfg=dict(type='BN', requires_grad=True),
+        norm_eval=True,
+        style='pytorch',
+        init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50')),
+    neck=dict(...),
+    rpn_head=dict(...),
+    roi_head=dict(...))
+```
+
+`ResNet` and `HRNet` use different keywords to construct.
+
+```python
+_base_ = '../mask_rcnn/mask-rcnn_r50_fpn_1x_coco.py'
+model = dict(
+    backbone=dict(
+        _delete_=True,
+        type='HRNet',
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(32, 64)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(32, 64, 128)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(32, 64, 128, 256))),
+        init_cfg=dict(type='Pretrained', checkpoint='open-mmlab://msra/hrnetv2_w32')),
+    neck=dict(...))
+```
+
+The `_delete_=True` would replace all old keys in `backbone` field with new keys.
+
+### Use intermediate variables in configs
+
+Some intermediate variables are used in the configs files, like `train_pipeline`/`test_pipeline` in datasets.
+It's worth noting that when modifying intermediate variables in the children configs, users need to pass the intermediate variables into corresponding fields again.
+For example, we would like to use a multi-scale strategy to train a Mask R-CNN. `train_pipeline`/`test_pipeline` are intermediate variables we would like to modify.
+
+```python
+_base_ = './mask-rcnn_r50_fpn_1x_coco.py'
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+    dict(
+        type='RandomResize', scale=[(1333, 640), (1333, 800)],
+        keep_ratio=True),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='PackDetInputs')
+]
+test_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='Resize', scale=(1333, 800), keep_ratio=True),
+    dict(
+        type='PackDetInputs',
+        meta_keys=('img_id', 'img_path', 'ori_shape', 'img_shape',
+                   'scale_factor'))
+]
+train_dataloader = dict(dataset=dict(pipeline=train_pipeline))
+val_dataloader = dict(dataset=dict(pipeline=test_pipeline))
+test_dataloader = dict(dataset=dict(pipeline=test_pipeline))
+```
+
+We first define the new `train_pipeline`/`test_pipeline` and pass them into dataloader fields.
+
+Similarly, if we would like to switch from `SyncBN` to `BN` or `MMSyncBN`, we need to substitute every `norm_cfg` in the config.
+
+```python
+_base_ = './mask-rcnn_r50_fpn_1x_coco.py'
+norm_cfg = dict(type='BN', requires_grad=True)
+model = dict(
+    backbone=dict(norm_cfg=norm_cfg),
+    neck=dict(norm_cfg=norm_cfg),
+    ...)
+```
+
+### Reuse variables in \_base\_ file
+
+If the users want to reuse the variables in the base file, they can get a copy of the corresponding variable by using `{{_base_.xxx}}`. E.g:
+
+```python
+_base_ = './mask-rcnn_r50_fpn_1x_coco.py'
+
+a = {{_base_.model}} # Variable `a` is equal to the `model` defined in `_base_`
+```
+
+## Modify config through script arguments
+
+When submitting jobs using `tools/train.py` or `tools/test.py`, you may specify `--cfg-options` to in-place modify the config.
+
+- Update config keys of dict chains.
+
+  The config options can be specified following the order of the dict keys in the original config.
+  For example, `--cfg-options model.backbone.norm_eval=False` changes the all BN modules in model backbones to `train` mode.
+
+- Update keys inside a list of configs.
+
+  Some config dicts are composed as a list in your config. For example, the training pipeline `train_dataloader.dataset.pipeline` is normally a list
+  e.g. `[dict(type='LoadImageFromFile'), ...]`. If you want to change `'LoadImageFromFile'` to `'LoadImageFromNDArray'` in the pipeline,
+  you may specify `--cfg-options data.train.pipeline.0.type=LoadImageFromNDArray`.
+
+- Update values of list/tuples.
+
+  If the value to be updated is a list or a tuple. For example, the config file normally sets `model.data_preprocessor.mean=[123.675, 116.28, 103.53]`. If you want to
+  change the mean values, you may specify `--cfg-options model.data_preprocessor.mean="[127,127,127]"`. Note that the quotation mark `"` is necessary to
+  support list/tuple data types, and **NO** white space is allowed inside the quotation marks in the specified value.
+
+## Config name style
+
+We follow the below style to name config files. Contributors are advised to follow the same style.
+
+```
+{algorithm name}_{model component names [component1]_[component2]_[...]}_{training settings}_{training dataset information}_{testing dataset information}.py
+```
+
+The file name is divided into five parts. All parts and components are connected with `_` and words of each part or component should be connected with `-`.
+
+- `{algorithm name}`: The name of the algorithm. It can be a detector name such as `faster-rcnn`, `mask-rcnn`, etc. Or can be a semi-supervised or knowledge-distillation algorithm such as `soft-teacher`, `lad`. etc.
+- `{model component names}`: Names of the components used in the algorithm such as backbone, neck, etc. For example, `r50-caffe_fpn_gn-head` means using caffe-style ResNet50, FPN and detection head with Group Norm in the algorithm.
+- `{training settings}`: Information of training settings such as batch size, augmentations, loss trick, scheduler, and epochs/iterations. For example: `4xb4-mixup-giou-coslr-100e` means using 8-gpus x 4-images-per-gpu, mixup augmentation, GIoU loss, cosine annealing learning rate, and train 100 epochs.
+  Some abbreviations:
+  - `{gpu x batch_per_gpu}`: GPUs and samples per GPU. `bN` indicates N batch size per GPU. E.g. `4xb4` is the short term of 4-GPUs x 4-images-per-GPU. And `8xb2` is used by default if not mentioned.
+  - `{schedule}`: training schedule, options are `1x`, `2x`, `20e`, etc.
+    `1x` and `2x` means 12 epochs and 24 epochs respectively.
+    `20e` is adopted in cascade models, which denotes 20 epochs.
+    For `1x`/`2x`, the initial learning rate decays by a factor of 10 at the 8/16th and 11/22th epochs.
+    For `20e`, the initial learning rate decays by a factor of 10 at the 16th and 19th epochs.
+- `{training dataset information}`: Training dataset names like `coco`, `coco-panoptic`, `cityscapes`, `voc-0712`, `wider-face`.
+- `{testing dataset information}` (optional): Testing dataset name for models trained on one dataset but tested on another. If not mentioned, it means the model was trained and tested on the same dataset type.

docs/en/user_guides/dataset_prepare.md

+# Dataset Prepare
+
+### Basic Detection Dataset Preparation
+
+MMDetection supports multiple public datasets including COCO, Pascal VOC, CityScapes, and [more](../../../configs/_base_/datasets).
+
+Public datasets like [Pascal VOC](http://host.robots.ox.ac.uk/pascal/VOC/index.html) or mirror and [COCO](https://cocodataset.org/#download) are available from official websites or mirrors. Note: In the detection task, Pascal VOC 2012 is an extension of Pascal VOC 2007 without overlap, and we usually use them together.
+It is recommended to download and extract the dataset somewhere outside the project directory and symlink the dataset root to `$MMDETECTION/data` as below.
+If your folder structure is different, you may need to change the corresponding paths in config files.
+
+We provide a script to download datasets such as COCO, you can run `python tools/misc/download_dataset.py --dataset-name coco2017` to download COCO dataset.
+For users in China, more datasets can be downloaded from the opensource dataset platform: [OpenDataLab](https://opendatalab.com/?source=OpenMMLab%20GitHub).
+
+For more usage please refer to [dataset-download](./useful_tools.md#dataset-download)
+
+```text
+mmdetection
+├── mmdet
+├── tools
+├── configs
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   ├── train2017
+│   │   ├── val2017
+│   │   ├── test2017
+│   ├── cityscapes
+│   │   ├── annotations
+│   │   ├── leftImg8bit
+│   │   │   ├── train
+│   │   │   ├── val
+│   │   ├── gtFine
+│   │   │   ├── train
+│   │   │   ├── val
+│   ├── VOCdevkit
+│   │   ├── VOC2007
+│   │   ├── VOC2012
+```
+
+Some models require additional [COCO-stuff](http://calvin.inf.ed.ac.uk/wp-content/uploads/data/cocostuffdataset/stuffthingmaps_trainval2017.zip) datasets, such as HTC, DetectoRS and SCNet, you can download, unzip, and then move them to the coco folder. The directory should be like this.
+
+```text
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   ├── train2017
+│   │   ├── val2017
+│   │   ├── test2017
+│   │   ├── stuffthingmaps
+```
+
+Panoptic segmentation models like PanopticFPN require additional [COCO Panoptic](http://images.cocodataset.org/annotations/panoptic_annotations_trainval2017.zip) datasets, you can download, unzip, and then move them to the coco annotation folder. The directory should be like this.
+
+```text
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   │   ├── panoptic_train2017.json
+│   │   │   ├── panoptic_train2017
+│   │   │   ├── panoptic_val2017.json
+│   │   │   ├── panoptic_val2017
+│   │   ├── train2017
+│   │   ├── val2017
+│   │   ├── test2017
+```
+
+The [cityscapes](https://www.cityscapes-dataset.com/) annotations need to be converted into the coco format using `tools/dataset_converters/cityscapes.py`:
+
+```shell
+pip install cityscapesscripts
+
+python tools/dataset_converters/cityscapes.py \
+    ./data/cityscapes \
+    --nproc 8 \
+    --out-dir ./data/cityscapes/annotations
+```
+
+### COCO Caption Dataset Preparation
+
+COCO Caption uses the COCO2014 dataset image and uses the annotation of karpathy.
+
+At first, you need to download the COCO2014 dataset.
+
+```shell
+python tools/misc/download_dataset.py --dataset-name coco2014 --unzip
+```
+
+The dataset will be downloaded to `data/coco` under the current path. Then download the annotation of karpathy.
+
+```shell
+cd data/coco/annotations
+wget https://storage.googleapis.com/sfr-vision-language-research/datasets/coco_karpathy_train.json
+wget https://storage.googleapis.com/sfr-vision-language-research/datasets/coco_karpathy_val.json
+wget https://storage.googleapis.com/sfr-vision-language-research/datasets/coco_karpathy_test.json
+wget https://storage.googleapis.com/sfr-vision-language-research/datasets/coco_karpathy_val_gt.json
+wget https://storage.googleapis.com/sfr-vision-language-research/datasets/coco_karpathy_test_gt.json
+```
+
+The final directory structure of the dataset folder that can be directly used for training and testing is as follows:
+
+```text
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   │   ├── coco_karpathy_train.json
+│   │   │   ├── coco_karpathy_test.json
+│   │   │   ├── coco_karpathy_val.json
+│   │   │   ├── coco_karpathy_val_gt.json
+│   │   │   ├── coco_karpathy_test_gt.json
+│   │   ├── train2014
+│   │   ├── val2014
+│   │   ├── test2014
+```
+
+### COCO Semantic Dataset Preparation
+
+There are two types of annotations for COCO semantic segmentation, which differ mainly in the definition of category names, so there are two ways to handle them. The first is to directly use the stuffthingmaps dataset, and the second is to use the panoptic dataset.
+
+**(1) Use stuffthingmaps dataset**
+
+The download link for this dataset is [stuffthingmaps_trainval2017](http://calvin.inf.ed.ac.uk/wp-content/uploads/data/cocostuffdataset/stuffthingmaps_trainval2017.zip). Please download and extract it to the `data/coco` folder.
+
+```text
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   ├── train2017
+│   │   ├── val2017
+│   │   ├── test2017
+│   │   ├── stuffthingmaps
+```
+
+This dataset is different from the standard COCO category annotation in that it includes 172 classes: 80 "thing" classes, 91 "stuff" classes, and 1 "unlabeled" class. The description of each class can be found at https://github.com/nightrome/cocostuff/blob/master/labels.md.
+
+Although only 172 categories are annotated, the maximum label ID in `stuffthingmaps` is 182, and some categories in the middle are not annotated. In addition, the "unlabeled" category of class 0 is removed. Therefore, the relationship between the value at each position in the final `stuffthingmaps` image can be found at https://github.com/kazuto1011/deeplab-pytorch/blob/master/data/datasets/cocostuff/labels.txt.
+
+To train efficiently and conveniently for users, we need to remove 12 unannotated classes before starting training or evaluation. The names of these 12 classes are: `street sign, hat, shoe, eye glasses, plate, mirror, window, desk, door, blender, hair brush`. The category information that can be used for training and evaluation can be found in `mmdet/datasets/coco_semantic.py`.
+
+You can use `tools/dataset_converters/coco_stuff164k.py` to convert the downloaded `stuffthingmaps` to a dataset that can be directly used for training and evaluation. The directory structure of the converted dataset is as follows:
+
+```text
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   ├── train2017
+│   │   ├── val2017
+│   │   ├── test2017
+│   │   ├── stuffthingmaps
+│   │   ├── stuffthingmaps_semseg
+```
+
+`stuffthingmaps_semseg` is the newly generated COCO semantic segmentation dataset that can be directly used for training and testing.
+
+**(2) use panoptic dataset**
+
+The number of categories in the semantic segmentation dataset generated through panoptic annotation will be less than that generated using the `stuffthingmaps` dataset. First, you need to prepare the panoptic segmentation annotations, and then use the following script to complete the conversion.
+
+```shell
+python tools/dataset_converters/prepare_coco_semantic_annos_from_panoptic_annos.py data/coco
+```
+
+The directory structure of the converted dataset is as follows:
+
+```text
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   │   ├── panoptic_train2017.json
+│   │   │   ├── panoptic_train2017
+│   │   │   ├── panoptic_val2017.json
+│   │   │   ├── panoptic_val2017
+│   │   │   ├── panoptic_semseg_train2017
+│   │   │   ├── panoptic_semseg_val2017
+│   │   ├── train2017
+│   │   ├── val2017
+│   │   ├── test2017
+```
+
+`panoptic_semseg_train2017` and `panoptic_semseg_val2017` are the newly generated COCO semantic segmentation datasets that can be directly used for training and testing. Note that their category information is the same as that of COCO panoptic segmentation, including both "thing" and "stuff" categories.
+
+### RefCOCO Dataset Preparation
+
+The images and annotations of [RefCOCO](https://github.com/lichengunc/refer) series datasets can be download by running `tools/misc/download_dataset.py`:
+
+```shell
+python tools/misc/download_dataset.py --dataset-name refcoco --save-dir data/coco --unzip
+```
+
+Then the directory should be like this:
+
+```text
+data
+├── coco
+│   ├── refcoco
+│   │   ├── instances.json
+│   │   ├── refs(google).p
+│   │   └── refs(unc).p
+│   ├── refcoco+
+│   │   ├── instances.json
+│   │   └── refs(unc).p
+│   ├── refcocog
+│   │   ├── instances.json
+│   │   ├── refs(google).p
+│   │   └── refs(umd).p
+│   │── train2014
+```
+
+### ADE20K 2016 Dataset Preparation
+
+The images and annotations of [ADE20K](https://groups.csail.mit.edu/vision/datasets/ADE20K/) dataset can be download by running `tools/misc/download_dataset.py`:
+
+```shell
+python tools/misc/download_dataset.py --dataset-name ade20k_2016 --save-dir data --unzip
+```
+
+Then move the annotations to the `data/ADEChallengeData2016` directory and run the preprocess script to produce the coco format annotations:
+
+```shell
+mv data/annotations_instance data/ADEChallengeData2016/
+mv data/categoryMapping.txt data/ADEChallengeData2016/
+mv data/imgCatIds.json data/ADEChallengeData2016/
+python tools/dataset_converters/ade20k2coco.py data/ADEChallengeData2016 --task panoptic
+python tools/dataset_converters/ade20k2coco.py data/ADEChallengeData2016 --task instance
+```
+
+The directory should be like this.
+
+```text
+data
+├── ADEChallengeData2016
+│   ├── ade20k_instance_train.json
+│   ├── ade20k_instance_val.json
+│   ├── ade20k_panoptic_train
+│   │   ├── ADE_train_00000001.png
+│   │   ├── ADE_train_00000002.png
+│   │   ├── ...
+│   ├── ade20k_panoptic_train.json
+│   ├── ade20k_panoptic_val
+│   │   ├── ADE_val_00000001.png
+│   │   ├── ADE_val_00000002.png
+│   │   ├── ...
+│   ├── ade20k_panoptic_val.json
+│   ├── annotations
+│   │   ├── training
+│   │   │   ├── ADE_train_00000001.png
+│   │   │   ├── ADE_train_00000002.png
+│   │   │   ├── ...
+│   │   ├── validation
+│   │   │   ├── ADE_val_00000001.png
+│   │   │   ├── ADE_val_00000002.png
+│   │   │   ├── ...
+│   ├── annotations_instance
+│   │   ├── training
+│   │   │   ├── ADE_train_00000001.png
+│   │   │   ├── ADE_train_00000002.png
+│   │   │   ├── ...
+│   │   ├── validation
+│   │   │   ├── ADE_val_00000001.png
+│   │   │   ├── ADE_val_00000002.png
+│   │   │   ├── ...
+│   ├── categoryMapping.txt
+│   ├── images
+│   │   ├── training
+│   │   │   ├── ADE_train_00000001.jpg
+│   │   │   ├── ADE_train_00000002.jpg
+│   │   │   ├── ...
+│   │   ├── validation
+│   │   │   ├── ADE_val_00000001.jpg
+│   │   │   ├── ADE_val_00000002.jpg
+│   │   │   ├── ...
+│   ├── imgCatIds.json
+│   ├── objectInfo150.txt
+│   │── sceneCategories.txt
+```
+
+The above folders include all data of ADE20K's semantic segmentation, instance segmentation, and panoptic segmentation.
+
+### Download from OpenDataLab
+
+By using [OpenDataLab](https://opendatalab.com/), researchers can obtain free formatted datasets in various fields. Through the search function of the platform, researchers may address the dataset they look for quickly and easily. Using the formatted datasets from the platform, researchers can efficiently conduct tasks across datasets.
+
+Currently, MIM supports downloading VOC and COCO datasets from OpenDataLab with one command line. More datasets will be supported in the future. You can also directly download the datasets you need from the OpenDataLab platform and then convert them to the format required by MMDetection.
+
+If you use MIM to download, make sure that the version is greater than v0.3.8. You can use the following command to update:
+
+```Bash
+pip install -U openmim
+```
+
+```Bash
+# install OpenXLab CLI tools
+pip install -U openxlab
+# log in OpenXLab, registry
+openxlab login
+
+# download voc2007 and preprocess by MIM
+mim download mmdet --dataset voc2007
+
+# download voc2012 and preprocess by MIM
+mim download mmdet --dataset voc2012
+
+# download coco2017 and preprocess by MIM
+mim download mmdet --dataset coco2017
+```

docs/en/user_guides/deploy.md

+# Model Deployment
+
+The deployment of OpenMMLab codebases, including MMDetection, MMPretrain and so on are supported by [MMDeploy](https://github.com/open-mmlab/mmdeploy).
+The latest deployment guide for MMDetection can be found from [here](https://mmdeploy.readthedocs.io/en/dev-1.x/04-supported-codebases/mmdet.html).
+
+This tutorial is organized as follows:
+
+- [Installation](#installation)
+- [Convert model](#convert-model)
+- [Model specification](#model-specification)
+- [Model inference](#model-inference)
+  - [Backend model inference](#backend-model-inference)
+  - [SDK model inference](#sdk-model-inference)
+- [Supported models](#supported-models)
+
+## Installation
+
+Please follow the [guide](https://mmdetection.readthedocs.io/en/latest/get_started.html) to install mmdet. And then install mmdeploy from source by following [this](https://mmdeploy.readthedocs.io/en/1.x/get_started.html#installation) guide.
+
+```{note}
+If you install mmdeploy prebuilt package, please also clone its repository by 'git clone https://github.com/open-mmlab/mmdeploy.git --depth=1' to get the deployment config files.
+```
+
+## Convert model
+
+Suppose mmdetection and mmdeploy repositories are in the same directory, and the working directory is the root path of mmdetection.
+
+Take [Faster R-CNN](https://github.com/open-mmlab/mmdetection/blob/main/configs/faster_rcnn/faster-rcnn_r50_fpn_1x_coco.py) model as an example. You can download its checkpoint from [here](https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r50_fpn_1x_coco/faster_rcnn_r50_fpn_1x_coco_20200130-047c8118.pth), and then convert it to onnx model as follows:
+
+```python
+from mmdeploy.apis import torch2onnx
+from mmdeploy.backend.sdk.export_info import export2SDK
+
+img = 'demo/demo.jpg'
+work_dir = 'mmdeploy_models/mmdet/onnx'
+save_file = 'end2end.onnx'
+deploy_cfg = '../mmdeploy/configs/mmdet/detection/detection_onnxruntime_dynamic.py'
+model_cfg = 'configs/faster_rcnn/faster-rcnn_r50_fpn_1x_coco.py'
+model_checkpoint = 'faster_rcnn_r50_fpn_1x_coco_20200130-047c8118.pth'
+device = 'cpu'
+
+# 1. convert model to onnx
+torch2onnx(img, work_dir, save_file, deploy_cfg, model_cfg,
+           model_checkpoint, device)
+
+# 2. extract pipeline info for inference by MMDeploy SDK
+export2SDK(deploy_cfg, model_cfg, work_dir, pth=model_checkpoint,
+           device=device)
+```
+
+It is crucial to specify the correct deployment config during model conversion. MMDeploy has already provided builtin deployment config [files](https://github.com/open-mmlab/mmdeploy/tree/1.x/configs/mmdet) of all supported backends for mmdetection, under which the config file path follows the pattern:
+
+```
+{task}/{task}_{backend}-{precision}_{static | dynamic}_{shape}.py
+```
+
+- **{task}:** task in mmdetection.
+
+  There are two of them. One is `detection` and the other is `instance-seg`, indicating instance segmentation.
+
+  mmdet models like `RetinaNet`, `Faster R-CNN` and `DETR` and so on belongs to `detection` task. While `Mask R-CNN` is one of `instance-seg` models.
+
+  **DO REMEMBER TO USE** `detection/detection_*.py` deployment config file when trying to convert detection models and use `instance-seg/instance-seg_*.py` to deploy instance segmentation models.
+
+- **{backend}:** inference backend, such as onnxruntime, tensorrt, pplnn, ncnn, openvino, coreml etc.
+
+- **{precision}:** fp16, int8. When it's empty, it means fp32
+
+- **{static | dynamic}:** static shape or dynamic shape
+
+- **{shape}:** input shape or shape range of a model
+
+Therefore, in the above example, you can also convert `Faster R-CNN` to tensorrt-fp16 model by `detection_tensorrt-fp16_dynamic-320x320-1344x1344.py`.
+
+```{tip}
+When converting mmdet models to tensorrt models, --device should be set to "cuda"
+```
+
+## Model specification
+
+Before moving on to model inference chapter, let's know more about the converted model structure which is very important for model inference.
+
+The converted model locates in the working directory like `mmdeploy_models/mmdet/onnx` in the previous example. It includes:
+
+```
+mmdeploy_models/mmdet/onnx
+├── deploy.json
+├── detail.json
+├── end2end.onnx
+└── pipeline.json
+```
+
+in which,
+
+- **end2end.onnx**: backend model which can be inferred by ONNX Runtime
+- ***xxx*.json**: the necessary information for mmdeploy SDK
+
+The whole package **mmdeploy_models/mmdet/onnx** is defined as **mmdeploy SDK model**, i.e., **mmdeploy SDK model** includes both backend model and inference meta information.
+
+## Model inference
+
+### Backend model inference
+
+Take the previous converted `end2end.onnx` model as an example, you can use the following code to inference the model and visualize the results.
+
+```python
+from mmdeploy.apis.utils import build_task_processor
+from mmdeploy.utils import get_input_shape, load_config
+import torch
+
+deploy_cfg = '../mmdeploy/configs/mmdet/detection/detection_onnxruntime_dynamic.py'
+model_cfg = 'configs/faster_rcnn/faster-rcnn_r50_fpn_1x_coco.py'
+device = 'cpu'
+backend_model = ['mmdeploy_models/mmdet/onnx/end2end.onnx']
+image = 'demo/demo.jpg'
+
+# read deploy_cfg and model_cfg
+deploy_cfg, model_cfg = load_config(deploy_cfg, model_cfg)
+
+# build task and backend model
+task_processor = build_task_processor(model_cfg, deploy_cfg, device)
+model = task_processor.build_backend_model(backend_model)
+
+# process input image
+input_shape = get_input_shape(deploy_cfg)
+model_inputs, _ = task_processor.create_input(image, input_shape)
+
+# do model inference
+with torch.no_grad():
+    result = model.test_step(model_inputs)
+
+# visualize results
+task_processor.visualize(
+    image=image,
+    model=model,
+    result=result[0],
+    window_name='visualize',
+    output_file='output_detection.png')
+```
+
+### SDK model inference
+
+You can also perform SDK model inference like following,
+
+```python
+from mmdeploy_python import Detector
+import cv2
+
+img = cv2.imread('demo/demo.jpg')
+
+# create a detector
+detector = Detector(model_path='mmdeploy_models/mmdet/onnx',
+                    device_name='cpu', device_id=0)
+# perform inference
+bboxes, labels, masks = detector(img)
+
+# visualize inference result
+indices = [i for i in range(len(bboxes))]
+for index, bbox, label_id in zip(indices, bboxes, labels):
+    [left, top, right, bottom], score = bbox[0:4].astype(int), bbox[4]
+    if score < 0.3:
+        continue
+
+    cv2.rectangle(img, (left, top), (right, bottom), (0, 255, 0))
+
+cv2.imwrite('output_detection.png', img)
+```
+
+Besides python API, mmdeploy SDK also provides other FFI (Foreign Function Interface), such as C, C++, C#, Java and so on. You can learn their usage from [demos](https://github.com/open-mmlab/mmdeploy/tree/1.x/demo).
+
+## Supported models
+
+Please refer to [here](https://mmdeploy.readthedocs.io/en/1.x/04-supported-codebases/mmdet.html#supported-models) for the supported model list.

docs/en/user_guides/finetune.md

+# Finetuning Models
+
+Detectors pre-trained on the COCO dataset can serve as a good pre-trained model for other datasets, e.g., CityScapes and KITTI Dataset.
+This tutorial provides instructions for users to use the models provided in the [Model Zoo](../model_zoo.md) for other datasets to obtain better performance.
+
+There are two steps to finetune a model on a new dataset.
+
+- Add support for the new dataset following [Customize Datasets](../advanced_guides/customize_dataset.md).
+- Modify the configs as will be discussed in this tutorial.
+
+Take the finetuning process on Cityscapes Dataset as an example, the users need to modify five parts in the config.
+
+## Inherit base configs
+
+To release the burden and reduce bugs in writing the whole configs, MMDetection V3.0 support inheriting configs from multiple existing configs. To finetune a Mask RCNN model, the new config needs to inherit
+`_base_/models/mask-rcnn_r50_fpn.py` to build the basic structure of the model. To use the Cityscapes Dataset, the new config can also simply inherit `_base_/datasets/cityscapes_instance.py`. For runtime settings such as logger settings, the new config needs to inherit `_base_/default_runtime.py`. For training schedules, the new config can to inherit `_base_/schedules/schedule_1x.py`. These configs are in the `configs` directory and the users can also choose to write the whole contents rather than use inheritance.
+
+```python
+_base_ = [
+    '../_base_/models/mask-rcnn_r50_fpn.py',
+    '../_base_/datasets/cityscapes_instance.py', '../_base_/default_runtime.py',
+    '../_base_/schedules/schedule_1x.py'
+]
+```
+
+## Modify head
+
+Then the new config needs to modify the head according to the class numbers of the new datasets. By only changing `num_classes` in the roi_head, the weights of the pre-trained models are mostly reused except for the final prediction head.
+
+```python
+model = dict(
+    roi_head=dict(
+        bbox_head=dict(
+            type='Shared2FCBBoxHead',
+            in_channels=256,
+            fc_out_channels=1024,
+            roi_feat_size=7,
+            num_classes=8,
+            bbox_coder=dict(
+                type='DeltaXYWHBBoxCoder',
+                target_means=[0., 0., 0., 0.],
+                target_stds=[0.1, 0.1, 0.2, 0.2]),
+            reg_class_agnostic=False,
+            loss_cls=dict(
+                type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
+            loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)),
+        mask_head=dict(
+            type='FCNMaskHead',
+            num_convs=4,
+            in_channels=256,
+            conv_out_channels=256,
+            num_classes=8,
+            loss_mask=dict(
+                type='CrossEntropyLoss', use_mask=True, loss_weight=1.0))))
+```
+
+## Modify dataset
+
+The users may also need to prepare the dataset and write the configs about dataset, refer to [Customize Datasets](../advanced_guides/customize_dataset.md) for more detail. MMDetection V3.0 already supports VOC, WIDERFACE, COCO, LIVS, OpenImages, DeepFashion, Objects365, and Cityscapes Dataset.
+
+## Modify training schedule
+
+The finetuning hyperparameters vary from the default schedule. It usually requires a smaller learning rate and fewer training epochs
+
+```python
+# optimizer
+# lr is set for a batch size of 8
+optim_wrapper = dict(optimizer=dict(lr=0.01))
+
+# learning rate
+param_scheduler = [
+    dict(
+        type='LinearLR', start_factor=0.001, by_epoch=False, begin=0, end=500),
+    dict(
+        type='MultiStepLR',
+        begin=0,
+        end=8,
+        by_epoch=True,
+        milestones=[7],
+        gamma=0.1)
+]
+
+# max_epochs
+train_cfg = dict(max_epochs=8)
+
+# log config
+default_hooks = dict(logger=dict(interval=100)),
+```
+
+## Use pre-trained model
+
+To use the pre-trained model, the new config adds the link of pre-trained models in the `load_from`. The users might need to download the model weights before training to avoid the download time during training.
+
+```python
+load_from = 'https://download.openmmlab.com/mmdetection/v2.0/mask_rcnn/mask_rcnn_r50_caffe_fpn_mstrain-poly_3x_coco/mask_rcnn_r50_caffe_fpn_mstrain-poly_3x_coco_bbox_mAP-0.408__segm_mAP-0.37_20200504_163245-42aa3d00.pth'  # noqa
+```

docs/en/user_guides/index.rst

+Train & Test
+**************
+
+MMDetection provides hundreds of pretrained detection models in `Model Zoo <https://mmdetection.readthedocs.io/en/latest/model_zoo.html>`_,
+and supports multiple standard datasets, including Pascal VOC, COCO, CityScapes, LVIS, etc. This note will show how to perform common tasks on these existing models and standard datasets:
+
+
+.. toctree::
+   :maxdepth: 1
+
+   config.md
+   inference.md
+   dataset_prepare.md
+   test.md
+   train.md
+   new_model.md
+   finetune.md
+   test_results_submission.md
+   init_cfg.md
+   single_stage_as_rpn.md
+   semi_det.md
+
+
+Useful Tools
+************
+
+.. toctree::
+   :maxdepth: 1
+
+   useful_tools.md
+   useful_hooks.md
+   visualization.md
+   robustness_benchmarking.md
+   deploy.md
+   label_studio.md
+   tracking_analysis_tools.md
+   tracking_config.md
+   tracking_dataset_prepare.md
+   tracking_inference.md
+   tracking_train_test.md
+   tracking_visualization.md

docs/en/user_guides/inference.md

+# Inference with existing models
+
+MMDetection provides hundreds of pre-trained detection models in [Model Zoo](https://mmdetection.readthedocs.io/en/latest/model_zoo.html).
+This note will show how to inference, which means using trained models to detect objects on images.
+
+In MMDetection, a model is defined by a [configuration file](https://mmdetection.readthedocs.io/en/latest/user_guides/config.html) and existing model parameters are saved in a checkpoint file.
+
+To start with, we recommend [RTMDet](https://github.com/open-mmlab/mmdetection/tree/main/configs/rtmdet) with this [configuration file](https://github.com/open-mmlab/mmdetection/blob/main/configs/rtmdet/rtmdet_l_8xb32-300e_coco.py) and this [checkpoint file](https://download.openmmlab.com/mmdetection/v3.0/rtmdet/rtmdet_l_8xb32-300e_coco/rtmdet_l_8xb32-300e_coco_20220719_112030-5a0be7c4.pth). It is recommended to download the checkpoint file to `checkpoints` directory.
+
+## High-level APIs for inference - `Inferencer`
+
+In OpenMMLab, all the inference operations are unified into a new interface - Inferencer. Inferencer is designed to expose a neat and simple API to users, and shares very similar interface across different OpenMMLab libraries.
+A notebook demo can be found in [demo/inference_demo.ipynb](https://github.com/open-mmlab/mmdetection/blob/main/demo/inference_demo.ipynb).
+
+### Basic Usage
+
+You can get inference results for an image with only 3 lines of code.
+
+```python
+from mmdet.apis import DetInferencer
+
+# Initialize the DetInferencer
+inferencer = DetInferencer('rtmdet_tiny_8xb32-300e_coco')
+
+# Perform inference
+inferencer('demo/demo.jpg', show=True)
+```
+
+The resulting output will be displayed in a new window:.
+
+<div align="center">
+    <img src='https://github.com/open-mmlab/mmdetection/assets/27466624/311df42d-640a-4a5b-9ad9-9ba7f3ec3a2f' />
+</div>
+
+```{note}
+If you are running MMDetection on a server without GUI or via SSH tunnel with X11 forwarding disabled, the `show` option will not work. However, you can still save visualizations to files by setting `out_dir` arguments. Read [Dumping Results](#dumping-results) for details.
+```
+
+### Initialization
+
+Each Inferencer must be initialized with a model. You can also choose the inference device during initialization.
+
+#### Model Initialization
+
+- To infer with MMDetection's pre-trained model, passing its name to the argument `model` can work. The weights will be automatically downloaded and loaded from OpenMMLab's model zoo.
+
+  ```python
+  inferencer = DetInferencer(model='rtmdet_tiny_8xb32-300e_coco')
+  ```
+
+  There is a very easy to list all model names in MMDetection.
+
+  ```python
+  # models is a list of model names, and them will print automatically
+  models = DetInferencer.list_models('mmdet')
+  ```
+
+  You can load another weight by passing its path/url to `weights`.
+
+  ```python
+  inferencer = DetInferencer(model='rtmdet_tiny_8xb32-300e_coco', weights='path/to/rtmdet.pth')
+  ```
+
+- To load custom config and weight, you can pass the path to the config file to `model` and the path to the weight to `weights`.
+
+  ```python
+  inferencer = DetInferencer(model='path/to/rtmdet_config.py', weights='path/to/rtmdet.pth')
+  ```
+
+- By default, [MMEngine](https://github.com/open-mmlab/mmengine/) dumps config to the weight. If you have a weight trained on MMEngine, you can also pass the path to the weight file to `weights` without specifying `model`:
+
+  ```python
+  # It will raise an error if the config file cannot be found in the weight. Currently, within the MMDetection model repository, only the weights of ddq-detr-4scale_r50 can be loaded in this manner.
+  inferencer = DetInferencer(weights='https://download.openmmlab.com/mmdetection/v3.0/ddq/ddq-detr-4scale_r50_8xb2-12e_coco/ddq-detr-4scale_r50_8xb2-12e_coco_20230809_170711-42528127.pth')
+  ```
+
+- Passing config file to `model` without specifying `weight` will result in a randomly initialized model.
+
+### Device
+
+Each Inferencer instance is bound to a device.
+By default, the best device is automatically decided by [MMEngine](https://github.com/open-mmlab/mmengine/). You can also alter the device by specifying the `device` argument. For example, you can use the following code to create an Inferencer on GPU 1.
+
+```python
+inferencer = DetInferencer(model='rtmdet_tiny_8xb32-300e_coco', device='cuda:1')
+```
+
+To create an Inferencer on CPU:
+
+```python
+inferencer = DetInferencer(model='rtmdet_tiny_8xb32-300e_coco', device='cpu')
+```
+
+Refer to [torch.device](https://pytorch.org/docs/stable/tensor_attributes.html#torch.device) for all the supported forms.
+
+### Inference
+
+Once the Inferencer is initialized, you can directly pass in the raw data to be inferred and get the inference results from return values.
+
+#### Input
+
+Input can be either of these types:
+
+- str: Path/URL to the image.
+
+  ```python
+  inferencer('demo/demo.jpg')
+  ```
+
+- array: Image in numpy array. It should be in BGR order.
+
+  ```python
+  import mmcv
+  array = mmcv.imread('demo/demo.jpg')
+  inferencer(array)
+  ```
+
+- list: A list of basic types above. Each element in the list will be processed separately.
+
+  ```python
+  inferencer(['img_1.jpg', 'img_2.jpg])
+  # You can even mix the types
+  inferencer(['img_1.jpg', array])
+  ```
+
+- str: Path to the directory. All images in the directory will be processed.
+
+  ```python
+  inferencer('path/to/your_imgs/')
+  ```
+
+### Output
+
+By default, each `Inferencer` returns the prediction results in a dictionary format.
+
+- `visualization` contains the visualized predictions.
+
+- `predictions` contains the predictions results in a json-serializable format. But it's an empty list by default unless `return_vis=True`.
+
+```python
+{
+      'predictions' : [
+        # Each instance corresponds to an input image
+        {
+          'labels': [...],  # int list of length (N, )
+          'scores': [...],  # float list of length (N, )
+          'bboxes': [...],  # 2d list of shape (N, 4), format: [min_x, min_y, max_x, max_y]
+        },
+        ...
+      ],
+      'visualization' : [
+        array(..., dtype=uint8),
+      ]
+  }
+```
+
+If you wish to get the raw outputs from the model, you can set `return_datasamples` to `True` to get the original [DataSample](advanced_guides/structures.md), which will be stored in `predictions`.
+
+#### Dumping Results
+
+Apart from obtaining predictions from the return value, you can also export the predictions/visualizations to files by setting `out_dir` and `no_save_pred`/`no_save_vis` arguments.
+
+```python
+inferencer('demo/demo.jpg', out_dir='outputs/', no_save_pred=False)
+```
+
+Results in the directory structure like:
+
+```text
+outputs
+├── preds
+│   └── demo.json
+└── vis
+    └── demo.jpg
+```
+
+The filename of each file is the same as the corresponding input image filename. If the input image is an array, the filename will be a number starting from 0.
+
+#### Batch Inference
+
+You can customize the batch size by setting `batch_size`. The default batch size is 1.
+
+### API
+
+Here are extensive lists of parameters that you can use.
+
+- **DetInferencer.\_\_init\_\_():**
+
+| Arguments       | Type          | Type    | Description                                                                                                                                                                                                                                                                                              |
+| --------------- | ------------- | ------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `model`         | str, optional | None    | Path to the config file or the model name defined in metafile. For example, it could be 'rtmdet-s' or 'rtmdet_s_8xb32-300e_coco' or 'configs/rtmdet/rtmdet_s_8xb32-300e_coco.py'. If the model is not specified, the user must provide the `weights` saved by MMEngine which contains the config string. |
+| `weights`       | str, optional | None    | Path to the checkpoint. If it is not specified and `model` is a model name of metafile, the weights will be loaded from metafile.                                                                                                                                                                        |
+| `device`        | str, optional | None    | Device used for inference, accepting all allowed strings by `torch.device`. E.g., 'cuda:0' or 'cpu'. If None, the available device will be automatically used.                                                                                                                                           |
+| `scope`         | str, optional | 'mmdet' | The scope of the model.                                                                                                                                                                                                                                                                                  |
+| `palette`       | str           | 'none'  | Color palette used for visualization. The order of priority is palette -> config -> checkpoint.                                                                                                                                                                                                          |
+| `show_progress` | bool          | True    | Control whether to display the progress bar during the inference process.                                                                                                                                                                                                                                |
+
+- **DetInferencer.\_\_call\_\_()**
+
+| Arguments            | Type                      | Default      | Description                                                                                                                                                                                                                                                    |
+| -------------------- | ------------------------- | ------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `inputs`             | str/list/tuple/np.array   | **required** | It can be a path to an image/a folder, an np array or a list/tuple (with img paths or np arrays)                                                                                                                                                               |
+| `batch_size`         | int                       | 1            | Inference batch size.                                                                                                                                                                                                                                          |
+| `print_result`       | bool                      | False        | Whether to print the inference result to the console.                                                                                                                                                                                                          |
+| `show`               | bool                      | False        | Whether to display the visualization results in a popup window.                                                                                                                                                                                                |
+| `wait_time`          | float                     | 0            | The interval of show(s).                                                                                                                                                                                                                                       |
+| `no_save_vis`        | bool                      | False        | Whether to force not to save prediction vis results.                                                                                                                                                                                                           |
+| `draw_pred`          | bool                      | True         | Whether to draw predicted bounding boxes.                                                                                                                                                                                                                      |
+| `pred_score_thr`     | float                     | 0.3          | Minimum score of bboxes to draw.                                                                                                                                                                                                                               |
+| `return_datasamples` | bool                      | False        | Whether to return results as DataSamples. If False, the results will be packed into a dict.                                                                                                                                                                    |
+| `print_result`       | bool                      | False        | Whether to print the inference result to the console.                                                                                                                                                                                                          |
+| `no_save_pred`       | bool                      | True         | Whether to force not to save prediction results.                                                                                                                                                                                                               |
+| `out_dir`            | str                       | ''           | Output directory of results.                                                                                                                                                                                                                                   |
+| `texts`              | str/list\[str\], optional | None         | Text prompts.                                                                                                                                                                                                                                                  |
+| `stuff_texts`        | str/list\[str\], optional | None         | Stuff text prompts of open panoptic task.                                                                                                                                                                                                                      |
+| `custom_entities`    | bool                      | False        | Whether to use custom entities. Only used in GLIP.                                                                                                                                                                                                             |
+| \*\*kwargs           |                           |              | Other keyword arguments passed to :meth:`preprocess`, :meth:`forward`, :meth:`visualize` and :meth:`postprocess`. Each key in kwargs should be in the corresponding set of `preprocess_kwargs`, `forward_kwargs`, `visualize_kwargs` and `postprocess_kwargs`. |
+
+## Demos
+
+We also provide four demo scripts, implemented with high-level APIs and supporting functionality codes.
+Source codes are available [here](https://github.com/open-mmlab/mmdetection/blob/main/demo).
+
+### Image demo
+
+This script performs inference on a single image.
+
+```shell
+python demo/image_demo.py \
+    ${IMAGE_FILE} \
+    ${CONFIG_FILE} \
+    [--weights ${WEIGHTS}] \
+    [--device ${GPU_ID}] \
+    [--pred-score-thr ${SCORE_THR}]
+```
+
+Examples:
+
+```shell
+python demo/image_demo.py demo/demo.jpg \
+    configs/rtmdet/rtmdet_l_8xb32-300e_coco.py \
+    --weights checkpoints/rtmdet_l_8xb32-300e_coco_20220719_112030-5a0be7c4.pth \
+    --device cpu
+```
+
+### Webcam demo
+
+This is a live demo from a webcam.
+
+```shell
+python demo/webcam_demo.py \
+    ${CONFIG_FILE} \
+    ${CHECKPOINT_FILE} \
+    [--device ${GPU_ID}] \
+    [--camera-id ${CAMERA-ID}] \
+    [--score-thr ${SCORE_THR}]
+```
+
+Examples:
+
+```shell
+python demo/webcam_demo.py \
+    configs/rtmdet/rtmdet_l_8xb32-300e_coco.py \
+    checkpoints/rtmdet_l_8xb32-300e_coco_20220719_112030-5a0be7c4.pth
+```
+
+### Video demo
+
+This script performs inference on a video.
+
+```shell
+python demo/video_demo.py \
+    ${VIDEO_FILE} \
+    ${CONFIG_FILE} \
+    ${CHECKPOINT_FILE} \
+    [--device ${GPU_ID}] \
+    [--score-thr ${SCORE_THR}] \
+    [--out ${OUT_FILE}] \
+    [--show] \
+    [--wait-time ${WAIT_TIME}]
+```
+
+Examples:
+
+```shell
+python demo/video_demo.py demo/demo.mp4 \
+    configs/rtmdet/rtmdet_l_8xb32-300e_coco.py \
+    checkpoints/rtmdet_l_8xb32-300e_coco_20220719_112030-5a0be7c4.pth \
+    --out result.mp4
+```
+
+#### Video demo with GPU acceleration
+
+This script performs inference on a video with GPU acceleration.
+
+```shell
+python demo/video_gpuaccel_demo.py \
+    ${VIDEO_FILE} \
+    ${CONFIG_FILE} \
+    ${CHECKPOINT_FILE} \
+    [--device ${GPU_ID}] \
+    [--score-thr ${SCORE_THR}] \
+    [--nvdecode] \
+    [--out ${OUT_FILE}] \
+    [--show] \
+    [--wait-time ${WAIT_TIME}]
+```
+
+Examples:
+
+```shell
+python demo/video_gpuaccel_demo.py demo/demo.mp4 \
+    configs/rtmdet/rtmdet_l_8xb32-300e_coco.py \
+    checkpoints/rtmdet_l_8xb32-300e_coco_20220719_112030-5a0be7c4.pth \
+    --nvdecode --out result.mp4
+```
+
+### Large-image inference demo
+
+This is a script for slicing inference on large images.
+
+```
+python demo/large_image_demo.py \
+	${IMG_PATH} \
+	${CONFIG_FILE} \
+	${CHECKPOINT_FILE} \
+	--device ${GPU_ID}  \
+	--show \
+	--tta  \
+	--score-thr ${SCORE_THR} \
+	--patch-size ${PATCH_SIZE} \
+	--patch-overlap-ratio ${PATCH_OVERLAP_RATIO} \
+	--merge-iou-thr ${MERGE_IOU_THR} \
+	--merge-nms-type ${MERGE_NMS_TYPE} \
+	--batch-size ${BATCH_SIZE} \
+	--debug \
+	--save-patch
+```
+
+Examples:
+
+```shell
+# inferecnce without tta
+wget -P checkpoint https://download.openmmlab.com/mmdetection/v2.0/faster_rcnn/faster_rcnn_r101_fpn_2x_coco/faster_rcnn_r101_fpn_2x_coco_bbox_mAP-0.398_20200504_210455-1d2dac9c.pth
+
+python demo/large_image_demo.py \
+    demo/large_image.jpg \
+    configs/faster_rcnn/faster-rcnn_r101_fpn_2x_coco.py \
+    checkpoint/faster_rcnn_r101_fpn_2x_coco_bbox_mAP-0.398_20200504_210455-1d2dac9c.pth
+
+# inference with tta
+wget -P checkpoint https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_fpn_1x_coco/retinanet_r50_fpn_1x_coco_20200130-c2398f9e.pth
+
+python demo/large_image_demo.py \
+    demo/large_image.jpg \
+    configs/retinanet/retinanet_r50_fpn_1x_coco.py \
+    checkpoint/retinanet_r50_fpn_1x_coco_20200130-c2398f9e.pth --tta
+
+```
+
+## Multi-modal algorithm inference demo and evaluation
+
+As multimodal vision algorithms continue to evolve, MMDetection has also supported such algorithms. This section demonstrates how to use the demo and eval scripts corresponding to multimodal algorithms using the GLIP algorithm and model as the example. Moreover, MMDetection integrated a [gradio_demo project](../../../projects/gradio_demo/), which allows developers to quickly play with all image input tasks in MMDetection on their local devices. Check the [document](../../../projects/gradio_demo/README.md) for more details.
+
+### Preparation
+
+Please first make sure that you have the correct dependencies installed:
+
+```shell
+# if source
+pip install -r requirements/multimodal.txt
+
+# if wheel
+mim install mmdet[multimodal]
+```
+
+MMDetection has already implemented GLIP algorithms and provided the weights, you can download directly from urls:
+
+```shell
+cd mmdetection
+wget https://download.openmmlab.com/mmdetection/v3.0/glip/glip_tiny_a_mmdet-b3654169.pth
+```
+
+### Inference
+
+Once the model is successfully downloaded, you can use the `demo/image_demo.py` script to run the inference.
+
+```shell
+python demo/image_demo.py demo/demo.jpg glip_tiny_a_mmdet-b3654169.pth --texts bench
+```
+
+Demo result will be similar to this:
+
+<div align=center>
+<img src="https://user-images.githubusercontent.com/17425982/234547841-266476c8-f987-4832-8642-34357be621c6.png" height="300"/>
+</div>
+
+If users would like to detect multiple targets, please declare them in the format of `xx. xx` after the `--texts`.
+
+```shell
+python demo/image_demo.py demo/demo.jpg glip_tiny_a_mmdet-b3654169.pth --texts 'bench. car'
+```
+
+And the result will be like this one:
+
+<div align=center>
+<img src="https://user-images.githubusercontent.com/17425982/234548156-ef9bbc2e-7605-4867-abe6-048b8578893d.png" height="300"/>
+</div>
+
+You can also use a sentence as the input prompt for the `--texts` field, for example:
+
+```shell
+python demo/image_demo.py demo/demo.jpg glip_tiny_a_mmdet-b3654169.pth --texts 'There are a lot of cars here.'
+```
+
+The result will be similar to this:
+
+<div align=center>
+<img src="https://user-images.githubusercontent.com/17425982/234548490-d2e0a16d-1aad-4708-aea0-c829634fabbd.png" height="300"/>
+</div>
+
+### Evaluation
+
+The GLIP implementation in MMDetection does not have any performance degradation, our benchmark is as follows:
+
+| Model                   | official mAP | mmdet mAP |
+| ----------------------- | :----------: | :-------: |
+| glip_A_Swin_T_O365.yaml |     42.9     |   43.0    |
+| glip_Swin_T_O365.yaml   |     44.9     |   44.9    |
+| glip_Swin_L.yaml        |     51.4     |   51.3    |
+
+Users can use the test script we provided to run evaluation as well. Here is a basic example:
+
+```shell
+# 1 gpu
+python tools/test.py configs/glip/glip_atss_swin-t_fpn_dyhead_pretrain_obj365.py glip_tiny_a_mmdet-b3654169.pth
+
+# 8 GPU
+./tools/dist_test.sh configs/glip/glip_atss_swin-t_fpn_dyhead_pretrain_obj365.py glip_tiny_a_mmdet-b3654169.pth 8
+```

docs/en/user_guides/init_cfg.md

+# Weight initialization
+
+During training, a proper initialization strategy is beneficial to speeding up the training or obtaining a higher performance. [MMCV](https://github.com/open-mmlab/mmcv/blob/master/mmcv/cnn/utils/weight_init.py) provide some commonly used methods for initializing modules like `nn.Conv2d`. Model initialization in MMdetection mainly uses `init_cfg`. Users can initialize models with following two steps:
+
+1. Define `init_cfg` for a model or its components in `model_cfg`,  but `init_cfg` of children components have higher priority and will override `init_cfg` of parents modules.
+2. Build model as usual, but call `model.init_weights()` method explicitly, and model parameters will be initialized as configuration.
+
+The high-level workflow of initialization in MMdetection is :
+
+model_cfg(init_cfg) -> build_from_cfg -> model -> init_weight() -> initialize(self, self.init_cfg) -> children's init_weight()
+
+### Description
+
+It is dict or list\[dict\], and contains the following keys and values:
+
+- `type` (str), containing the initializer name in `INTIALIZERS`, and followed by arguments of the initializer.
+- `layer` (str or list\[str\]), containing the names of basic layers in Pytorch or MMCV with learnable parameters that will be initialized, e.g. `'Conv2d'`,`'DeformConv2d'`.
+- `override` (dict or list\[dict\]),  containing the sub-modules that not inherit from BaseModule and whose initialization configuration is different from other layers' which are in `'layer'` key. Initializer defined in `type` will work for all layers defined in `layer`, so if sub-modules are not derived Classes of `BaseModule` but can be initialized as same ways of layers in `layer`, it does not need to use `override`. `override` contains:
+  - `type` followed by arguments of initializer;
+  - `name` to indicate sub-module which will be initialized.
+
+### Initialize parameters
+
+Inherit a new model from `mmcv.runner.BaseModule` or `mmdet.models`  Here we show an example of FooModel.
+
+```python
+import torch.nn as nn
+from mmcv.runner import BaseModule
+
+class FooModel(BaseModule)
+	def __init__(self,
+                 arg1,
+                 arg2,
+                 init_cfg=None):
+    	super(FooModel, self).__init__(init_cfg)
+		...
+```
+
+- Initialize model by using `init_cfg` directly in code
+
+  ```python
+  import torch.nn as nn
+  from mmcv.runner import BaseModule
+  # or directly inherit mmdet models
+
+  class FooModel(BaseModule)
+  	def __init__(self,
+                  arg1,
+                  arg2,
+                  init_cfg=XXX):
+  		super(FooModel, self).__init__(init_cfg)
+  	    ...
+  ```
+
+- Initialize model by using `init_cfg` directly in `mmcv.Sequential` or `mmcv.ModuleList` code
+
+  ```python
+  from mmcv.runner import BaseModule, ModuleList
+
+  class FooModel(BaseModule)
+  	def __init__(self,
+              	arg1,
+              	arg2,
+              	init_cfg=None):
+  		super(FooModel, self).__init__(init_cfg)
+      	...
+      	self.conv1 = ModuleList(init_cfg=XXX)
+  ```
+
+- Initialize model by using `init_cfg` in config file
+
+  ```python
+  model = dict(
+  	...
+  	model = dict(
+      	type='FooModel',
+      	arg1=XXX,
+      	arg2=XXX,
+      	init_cfg=XXX),
+          ...
+  ```
+
+### Usage of init_cfg
+
+1. Initialize model by `layer` key
+
+   If we only define `layer`, it just initialize the layer in `layer` key.
+
+   NOTE: Value of `layer` key is the class name with attributes weights and bias of Pytorch, (so such as  `MultiheadAttention layer` is not supported).
+
+- Define `layer` key for initializing module with same configuration.
+
+  ```python
+  init_cfg = dict(type='Constant', layer=['Conv1d', 'Conv2d', 'Linear'], val=1)
+  # initialize whole module with same configuration
+  ```
+
+- Define `layer` key for initializing layer with different configurations.
+
+```python
+init_cfg = [dict(type='Constant', layer='Conv1d', val=1),
+            dict(type='Constant', layer='Conv2d', val=2),
+            dict(type='Constant', layer='Linear', val=3)]
+# nn.Conv1d will be initialized with dict(type='Constant', val=1)
+# nn.Conv2d will be initialized with dict(type='Constant', val=2)
+# nn.Linear will be initialized with dict(type='Constant', val=3)
+```
+
+2. Initialize model by `override` key
+
+- When initializing some specific part with its attribute name, we can use `override` key, and the value in `override` will ignore the value in init_cfg.
+
+  ```python
+  # layers:
+  # self.feat = nn.Conv1d(3, 1, 3)
+  # self.reg = nn.Conv2d(3, 3, 3)
+  # self.cls = nn.Linear(1,2)
+
+  init_cfg = dict(type='Constant',
+                  layer=['Conv1d','Conv2d'], val=1, bias=2,
+                  override=dict(type='Constant', name='reg', val=3, bias=4))
+  # self.feat and self.cls will be initialized with 	dict(type='Constant', val=1, bias=2)
+  # The module called 'reg' will be initialized with dict(type='Constant', val=3, bias=4)
+  ```
+
+- If `layer` is None in init_cfg, only sub-module with the name in override will be initialized, and type and other args in override can be omitted.
+
+  ```python
+  # layers:
+  # self.feat = nn.Conv1d(3, 1, 3)
+  # self.reg = nn.Conv2d(3, 3, 3)
+  # self.cls = nn.Linear(1,2)
+
+  init_cfg = dict(type='Constant', val=1, bias=2, 	override=dict(name='reg'))
+
+  # self.feat and self.cls will be initialized by Pytorch
+  # The module called 'reg' will be initialized with dict(type='Constant', val=1, bias=2)
+  ```
+
+- If we don't define `layer` key or `override` key, it will not initialize anything.
+
+- Invalid usage
+
+  ```python
+  # It is invalid that override don't have name key
+  init_cfg = dict(type='Constant', layer=['Conv1d','Conv2d'], val=1, bias=2,
+              	override=dict(type='Constant', val=3, bias=4))
+
+  # It is also invalid that override has name and other args except type
+  init_cfg = dict(type='Constant', layer=['Conv1d','Conv2d'], val=1, bias=2,
+                  override=dict(name='reg', val=3, bias=4))
+  ```
+
+3. Initialize model with the pretrained model
+
+   ```python
+   init_cfg = dict(type='Pretrained',
+               checkpoint='torchvision://resnet50')
+   ```
+
+More details can refer to the documentation in [MMEngine](https://mmengine.readthedocs.io/en/latest/advanced_tutorials/initialize.html)

docs/en/user_guides/label_studio.md

+# Semi-automatic Object Detection Annotation with MMDetection and Label-Studio
+
+Annotation data is a time-consuming and laborious task. This article introduces how to perform semi-automatic annotation using the RTMDet algorithm in MMDetection in conjunction with Label-Studio software. Specifically, using RTMDet to predict image annotations and then refining the annotations with Label-Studio. Community users can refer to this process and methodology and apply it to other fields.
+
+- RTMDet: RTMDet is a high-precision single-stage object detection algorithm developed by OpenMMLab, open-sourced in the MMDetection object detection toolbox. Its open-source license is Apache 2.0, and it can be used freely without restrictions by industrial users.
+
+- [Label Studio](https://github.com/heartexlabs/label-studio) is an excellent annotation software covering the functionality of dataset annotation in areas such as image classification, object detection, and segmentation.
+
+In this article, we will use [cat](https://download.openmmlab.com/mmyolo/data/cat_dataset.zip) images for semi-automatic annotation.
+
+## Environment Configuration
+
+To begin with, you need to create a virtual environment and then install PyTorch and MMCV. In this article, we will specify the versions of PyTorch and MMCV. Next, you can install MMDetection, Label-Studio, and label-studio-ml-backend using the following steps:
+
+Create a virtual environment:
+
+```shell
+conda create -n rtmdet python=3.9 -y
+conda activate rtmdet
+```
+
+Install PyTorch:
+
+```shell
+# Linux and Windows CPU only
+pip install torch==1.10.1+cpu torchvision==0.11.2+cpu torchaudio==0.10.1 -f https://download.pytorch.org/whl/cpu/torch_stable.html
+# Linux and Windows CUDA 11.3
+pip install torch==1.10.1+cu113 torchvision==0.11.2+cu113 torchaudio==0.10.1 -f https://download.pytorch.org/whl/cu113/torch_stable.html
+# OSX
+pip install torch==1.10.1 torchvision==0.11.2 torchaudio==0.10.1
+```
+
+Install MMCV:
+
+```shell
+pip install -U openmim
+mim install "mmcv>=2.0.0"
+# Installing mmcv will automatically install mmengine
+```
+
+Install MMDetection:
+
+```shell
+git clone https://github.com/open-mmlab/mmdetection
+cd mmdetection
+pip install -v -e .
+```
+
+Install Label-Studio and label-studio-ml-backend:
+
+```shell
+# Installing Label-Studio may take some time, if the version is not found, please use the official source
+pip install label-studio==1.7.2
+pip install label-studio-ml==1.0.9
+```
+
+Download the rtmdet weights:
+
+```shell
+cd path/to/mmetection
+mkdir work_dirs
+cd work_dirs
+wget https://download.openmmlab.com/mmdetection/v3.0/rtmdet/rtmdet_m_8xb32-300e_coco/rtmdet_m_8xb32-300e_coco_20220719_112220-229f527c.pth
+```
+
+## Start the Service
+
+Start the RTMDet backend inference service:
+
+```shell
+cd path/to/mmetection
+
+label-studio-ml start projects/LabelStudio/backend_template --with \
+config_file=configs/rtmdet/rtmdet_m_8xb32-300e_coco.py \
+checkpoint_file=./work_dirs/rtmdet_m_8xb32-300e_coco_20220719_112220-229f527c.pth \
+device=cpu \
+--port 8003
+# Set device=cpu to use CPU inference, and replace cpu with cuda:0 to use GPU inference.
+```
+
+![](https://cdn.vansin.top/picgo20230330131601.png)
+
+The RTMDet backend inference service has now been started. To configure it in the Label-Studio web system, use http://localhost:8003 as the backend inference service.
+
+Now, start the Label-Studio web service:
+
+```shell
+label-studio start
+```
+
+![](https://cdn.vansin.top/picgo20230330132913.png)
+
+Open your web browser and go to http://localhost:8080/ to see the Label-Studio interface.
+
+![](https://cdn.vansin.top/picgo20230330133118.png)
+
+Register a user and then create an RTMDet-Semiautomatic-Label project.
+
+![](https://cdn.vansin.top/picgo20230330133333.png)
+
+Download the example cat images by running the following command and import them using the Data Import button:
+
+```shell
+cd path/to/mmetection
+mkdir data && cd data
+
+wget https://download.openmmlab.com/mmyolo/data/cat_dataset.zip && unzip cat_dataset.zip
+```
+
+![](https://cdn.vansin.top/picgo20230330133628.png)
+
+![](https://cdn.vansin.top/picgo20230330133715.png)
+
+Then, select the Object Detection With Bounding Boxes template.
+
+![](https://cdn.vansin.top/picgo20230330133807.png)
+
+```shell
+airplane
+apple
+backpack
+banana
+baseball_bat
+baseball_glove
+bear
+bed
+bench
+bicycle
+bird
+boat
+book
+bottle
+bowl
+broccoli
+bus
+cake
+car
+carrot
+cat
+cell_phone
+chair
+clock
+couch
+cow
+cup
+dining_table
+dog
+donut
+elephant
+fire_hydrant
+fork
+frisbee
+giraffe
+hair_drier
+handbag
+horse
+hot_dog
+keyboard
+kite
+knife
+laptop
+microwave
+motorcycle
+mouse
+orange
+oven
+parking_meter
+person
+pizza
+potted_plant
+refrigerator
+remote
+sandwich
+scissors
+sheep
+sink
+skateboard
+skis
+snowboard
+spoon
+sports_ball
+stop_sign
+suitcase
+surfboard
+teddy_bear
+tennis_racket
+tie
+toaster
+toilet
+toothbrush
+traffic_light
+train
+truck
+tv
+umbrella
+vase
+wine_glass
+zebra
+```
+
+Then, copy and add the above categories to Label-Studio and click Save.
+
+![](https://cdn.vansin.top/picgo20230330134027.png)
+
+In the Settings, click Add Model to add the RTMDet backend inference service.
+
+![](https://cdn.vansin.top/picgo20230330134320.png)
+
+Click Validate and Save, and then click Start Labeling.
+
+![](https://cdn.vansin.top/picgo20230330134424.png)
+
+If you see Connected as shown below, the backend inference service has been successfully added.
+
+![](https://cdn.vansin.top/picgo20230330134554.png)
+
+## Start Semi-Automatic Labeling
+
+Click on Label to start labeling.
+
+![](https://cdn.vansin.top/picgo20230330134804.png)
+
+We can see that the RTMDet backend inference service has successfully returned the predicted results and displayed them on the image. However, we noticed that the predicted bounding boxes for the cats are a bit too large and not very accurate.
+
+![](https://cdn.vansin.top/picgo20230403104419.png)
+
+We manually adjust the position of the cat bounding box, and then click Submit to complete the annotation of this image.
+
+![](https://cdn.vansin.top/picgo/20230403105923.png)
+
+After submitting all images, click export to export the labeled dataset in COCO format.
+
+![](https://cdn.vansin.top/picgo20230330135921.png)
+
+Use VS Code to open the unzipped folder to see the labeled dataset, which includes the images and the annotation files in JSON format.
+
+![](https://cdn.vansin.top/picgo20230330140321.png)
+
+At this point, the semi-automatic labeling is complete. We can use this dataset to train a more accurate model in MMDetection and then continue semi-automatic labeling on newly collected images with this model. This way, we can iteratively expand the high-quality dataset and improve the accuracy of the model.
+
+## Use MMYOLO as the Backend Inference Service
+
+If you want to use Label-Studio in MMYOLO, you can refer to replacing the config_file and checkpoint_file with the configuration file and weight file of MMYOLO when starting the backend inference service.
+
+```shell
+cd path/to/mmetection
+
+label-studio-ml start projects/LabelStudio/backend_template --with \
+config_file= path/to/mmyolo_config.py \
+checkpoint_file= path/to/mmyolo_weights.pth \
+device=cpu \
+--port 8003
+# device=cpu is for using CPU inference. If using GPU inference, replace cpu with cuda:0.
+```
+
+Rotation object detection and instance segmentation are still under development, please stay tuned.

docs/en/user_guides/new_model.md

+# Train with customized models and standard datasets
+
+In this note, you will know how to train, test and inference your own customized models under standard datasets. We use the cityscapes dataset to train a customized Cascade Mask R-CNN R50 model as an example to demonstrate the whole process, which using [`AugFPN`](https://github.com/Gus-Guo/AugFPN) to replace the default `FPN` as neck, and add `Rotate` or `TranslateX` as training-time auto augmentation.
+
+The basic steps are as below:
+
+1. Prepare the standard dataset
+2. Prepare your own customized model
+3. Prepare a config
+4. Train, test, and inference models on the standard dataset.
+
+## Prepare the standard dataset
+
+In this note, as we use the standard cityscapes dataset as an example.
+
+It is recommended to symlink the dataset root to `$MMDETECTION/data`.
+If your folder structure is different, you may need to change the corresponding paths in config files.
+
+```none
+mmdetection
+├── mmdet
+├── tools
+├── configs
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   ├── train2017
+│   │   ├── val2017
+│   │   ├── test2017
+│   ├── cityscapes
+│   │   ├── annotations
+│   │   ├── leftImg8bit
+│   │   │   ├── train
+│   │   │   ├── val
+│   │   ├── gtFine
+│   │   │   ├── train
+│   │   │   ├── val
+│   ├── VOCdevkit
+│   │   ├── VOC2007
+│   │   ├── VOC2012
+
+```
+
+Or you can set your dataset root through
+
+```bash
+export MMDET_DATASETS=$data_root
+```
+
+We will replace dataset root with `$MMDET_DATASETS`, so you don't have to modify the corresponding path in config files.
+
+The cityscapes annotations have to be converted into the coco format using `tools/dataset_converters/cityscapes.py`:
+
+```shell
+pip install cityscapesscripts
+python tools/dataset_converters/cityscapes.py ./data/cityscapes --nproc 8 --out-dir ./data/cityscapes/annotations
+```
+
+Currently, the config files in `cityscapes` use COCO pre-trained weights to initialize.
+You could download the pre-trained models in advance if the network is unavailable or slow, otherwise, it would cause errors at the beginning of training.
+
+## Prepare your own customized model
+
+The second step is to use your own module or training setting. Assume that we want to implement a new neck called `AugFPN` to replace with the default `FPN` under the existing detector Cascade Mask R-CNN R50. The following implements `AugFPN` under MMDetection.
+
+### 1. Define a new neck (e.g. AugFPN)
+
+Firstly create a new file `mmdet/models/necks/augfpn.py`.
+
+```python
+import torch.nn as nn
+from mmdet.registry import MODELS
+
+
+@MODELS.register_module()
+class AugFPN(nn.Module):
+
+    def __init__(self,
+                in_channels,
+                out_channels,
+                num_outs,
+                start_level=0,
+                end_level=-1,
+                add_extra_convs=False):
+        pass
+
+    def forward(self, inputs):
+        # implementation is ignored
+        pass
+```
+
+### 2. Import the module
+
+You can either add the following line to `mmdet/models/necks/__init__.py`,
+
+```python
+from .augfpn import AugFPN
+```
+
+or alternatively add
+
+```python
+custom_imports = dict(
+    imports=['mmdet.models.necks.augfpn'],
+    allow_failed_imports=False)
+```
+
+to the config file and avoid modifying the original code.
+
+### 3. Modify the config file
+
+```python
+neck=dict(
+    type='AugFPN',
+    in_channels=[256, 512, 1024, 2048],
+    out_channels=256,
+    num_outs=5)
+```
+
+For more detailed usages about customizing your own models (e.g. implement a new backbone, head, loss, etc) and runtime training settings (e.g. define a new optimizer, use gradient clip, customize training schedules and hooks, etc), please refer to the guideline [Customize Models](../advanced_guides/customize_models.md) and [Customize Runtime Settings](../advanced_guides/customize_runtime.md) respectively.
+
+## Prepare a config
+
+The third step is to prepare a config for your own training setting. Assume that we want to add `AugFPN` and `Rotate` or `Translate` augmentation to existing Cascade Mask R-CNN R50 to train the cityscapes dataset, and assume the config is under directory `configs/cityscapes/` and named as `cascade-mask-rcnn_r50_augfpn_autoaug-10e_cityscapes.py`, the config is as below.
+
+```python
+# The new config inherits the base configs to highlight the necessary modification
+_base_ = [
+    '../_base_/models/cascade-mask-rcnn_r50_fpn.py',
+    '../_base_/datasets/cityscapes_instance.py', '../_base_/default_runtime.py'
+]
+
+model = dict(
+    # set None to avoid loading ImageNet pre-trained backbone,
+    # instead here we set `load_from` to load from COCO pre-trained detectors.
+    backbone=dict(init_cfg=None),
+    # replace neck from defaultly `FPN` to our new implemented module `AugFPN`
+    neck=dict(
+        type='AugFPN',
+        in_channels=[256, 512, 1024, 2048],
+        out_channels=256,
+        num_outs=5),
+    # We also need to change the num_classes in head from 80 to 8, to match the
+    # cityscapes dataset's annotation. This modification involves `bbox_head` and `mask_head`.
+    roi_head=dict(
+        bbox_head=[
+            dict(
+                type='Shared2FCBBoxHead',
+                in_channels=256,
+                fc_out_channels=1024,
+                roi_feat_size=7,
+                # change the number of classes from defaultly COCO to cityscapes
+                num_classes=8,
+                bbox_coder=dict(
+                    type='DeltaXYWHBBoxCoder',
+                    target_means=[0., 0., 0., 0.],
+                    target_stds=[0.1, 0.1, 0.2, 0.2]),
+                reg_class_agnostic=True,
+                loss_cls=dict(
+                    type='CrossEntropyLoss',
+                    use_sigmoid=False,
+                    loss_weight=1.0),
+                loss_bbox=dict(type='SmoothL1Loss', beta=1.0,
+                               loss_weight=1.0)),
+            dict(
+                type='Shared2FCBBoxHead',
+                in_channels=256,
+                fc_out_channels=1024,
+                roi_feat_size=7,
+                # change the number of classes from defaultly COCO to cityscapes
+                num_classes=8,
+                bbox_coder=dict(
+                    type='DeltaXYWHBBoxCoder',
+                    target_means=[0., 0., 0., 0.],
+                    target_stds=[0.05, 0.05, 0.1, 0.1]),
+                reg_class_agnostic=True,
+                loss_cls=dict(
+                    type='CrossEntropyLoss',
+                    use_sigmoid=False,
+                    loss_weight=1.0),
+                loss_bbox=dict(type='SmoothL1Loss', beta=1.0,
+                               loss_weight=1.0)),
+            dict(
+                type='Shared2FCBBoxHead',
+                in_channels=256,
+                fc_out_channels=1024,
+                roi_feat_size=7,
+                # change the number of classes from defaultly COCO to cityscapes
+                num_classes=8,
+                bbox_coder=dict(
+                    type='DeltaXYWHBBoxCoder',
+                    target_means=[0., 0., 0., 0.],
+                    target_stds=[0.033, 0.033, 0.067, 0.067]),
+                reg_class_agnostic=True,
+                loss_cls=dict(
+                    type='CrossEntropyLoss',
+                    use_sigmoid=False,
+                    loss_weight=1.0),
+                loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0))
+        ],
+        mask_head=dict(
+            type='FCNMaskHead',
+            num_convs=4,
+            in_channels=256,
+            conv_out_channels=256,
+            # change the number of classes from default COCO to cityscapes
+            num_classes=8,
+            loss_mask=dict(
+                type='CrossEntropyLoss', use_mask=True, loss_weight=1.0))))
+
+# over-write `train_pipeline` for new added `AutoAugment` training setting
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
+    dict(
+        type='AutoAugment',
+        policies=[
+            [dict(
+                 type='Rotate',
+                 level=5,
+                 img_border_value=(124, 116, 104),
+                 prob=0.5)
+            ],
+            [dict(type='Rotate', level=7, img_border_value=(124, 116, 104)),
+             dict(
+                 type='TranslateX',
+                 level=5,
+                 prob=0.5,
+                 img_border_value=(124, 116, 104))
+            ],
+        ]),
+    dict(
+        type='RandomResize',
+        scale=[(2048, 800), (2048, 1024)],
+        keep_ratio=True),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='PackDetInputs'),
+]
+
+# set batch_size per gpu, and set new training pipeline
+train_dataloader = dict(
+    batch_size=1,
+    num_workers=3,
+    # over-write `pipeline` with new training pipeline setting
+    dataset=dict(pipeline=train_pipeline))
+
+# Set optimizer
+optim_wrapper = dict(
+    type='OptimWrapper',
+    optimizer=dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001))
+
+# Set customized learning policy
+param_scheduler = [
+    dict(
+        type='LinearLR', start_factor=0.001, by_epoch=False, begin=0, end=500),
+    dict(
+        type='MultiStepLR',
+        begin=0,
+        end=10,
+        by_epoch=True,
+        milestones=[8],
+        gamma=0.1)
+]
+
+# train, val, test loop config
+train_cfg = dict(max_epochs=10, val_interval=1)
+
+# We can use the COCO pre-trained Cascade Mask R-CNN R50 model for a more stable performance initialization
+load_from = 'https://download.openmmlab.com/mmdetection/v2.0/cascade_rcnn/cascade_mask_rcnn_r50_fpn_1x_coco/cascade_mask_rcnn_r50_fpn_1x_coco_20200203-9d4dcb24.pth'
+```
+
+## Train a new model
+
+To train a model with the new config, you can simply run
+
+```shell
+python tools/train.py configs/cityscapes/cascade-mask-rcnn_r50_augfpn_autoaug-10e_cityscapes.py
+```
+
+For more detailed usages, please refer to the [training guide](train.md).
+
+## Test and inference
+
+To test the trained model, you can simply run
+
+```shell
+python tools/test.py configs/cityscapes/cascade-mask-rcnn_r50_augfpn_autoaug-10e_cityscapes.py work_dirs/cascade-mask-rcnn_r50_augfpn_autoaug-10e_cityscapes/epoch_10.pth
+```
+
+For more detailed usages, please refer to the [testing guide](test.md).

docs/en/user_guides/robustness_benchmarking.md

+# Corruption Benchmarking
+
+## Introduction
+
+We provide tools to test object detection and instance segmentation models on the image corruption benchmark defined in [Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming](https://arxiv.org/abs/1907.07484).
+This page provides basic tutorials how to use the benchmark.
+
+```latex
+@article{michaelis2019winter,
+  title={Benchmarking Robustness in Object Detection:
+    Autonomous Driving when Winter is Coming},
+  author={Michaelis, Claudio and Mitzkus, Benjamin and
+    Geirhos, Robert and Rusak, Evgenia and
+    Bringmann, Oliver and Ecker, Alexander S. and
+    Bethge, Matthias and Brendel, Wieland},
+  journal={arXiv:1907.07484},
+  year={2019}
+}
+```
+
+![image corruption example](../../../resources/corruptions_sev_3.png)
+
+## About the benchmark
+
+To submit results to the benchmark please visit the [benchmark homepage](https://github.com/bethgelab/robust-detection-benchmark)
+
+The benchmark is modelled after the [imagenet-c benchmark](https://github.com/hendrycks/robustness) which was originally
+published in [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://arxiv.org/abs/1903.12261) (ICLR 2019) by Dan Hendrycks and Thomas Dietterich.
+
+The image corruption functions are included in this library but can be installed separately using:
+
+```shell
+pip install imagecorruptions
+```
+
+Compared to imagenet-c a few changes had to be made to handle images of arbitrary size and greyscale images.
+We also modified the 'motion blur' and 'snow' corruptions to remove dependency from a linux specific library,
+which would have to be installed separately otherwise. For details please refer to the [imagecorruptions repository](https://github.com/bethgelab/imagecorruptions).
+
+## Inference with pretrained models
+
+We provide a testing script to evaluate a models performance on any combination of the corruptions provided in the benchmark.
+
+### Test a dataset
+
+- [x] single GPU testing
+- [ ] multiple GPU testing
+- [ ] visualize detection results
+
+You can use the following commands to test a models performance under the 15 corruptions used in the benchmark.
+
+```shell
+# single-gpu testing
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}]
+```
+
+Alternatively different group of corruptions can be selected.
+
+```shell
+# noise
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] --corruptions noise
+
+# blur
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] --corruptions blur
+
+# wetaher
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] --corruptions weather
+
+# digital
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] --corruptions digital
+```
+
+Or a costom set of corruptions e.g.:
+
+```shell
+# gaussian noise, zoom blur and snow
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [--eval ${EVAL_METRICS}] --corruptions gaussian_noise zoom_blur snow
+```
+
+Finally the corruption severities to evaluate can be chosen.
+Severity 0 corresponds to clean data and the effect increases from 1 to 5.
+
+```shell
+# severity 1
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [--eval ${EVAL_METRICS}] --severities 1
+
+# severities 0,2,4
+python tools/analysis_tools/test_robustness.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [--eval ${EVAL_METRICS}] --severities 0 2 4
+```
+
+## Results for modelzoo models
+
+The results on COCO 2017val are shown in the below table.
+
+|        Model        |      Backbone       |  Style  | Lr schd | box AP clean | box AP corr. | box % | mask AP clean | mask AP corr. | mask % |
+| :-----------------: | :-----------------: | :-----: | :-----: | :----------: | :----------: | :---: | :-----------: | :-----------: | :----: |
+|    Faster R-CNN     |      R-50-FPN       | pytorch |   1x    |     36.3     |     18.2     | 50.2  |       -       |       -       |   -    |
+|    Faster R-CNN     |      R-101-FPN      | pytorch |   1x    |     38.5     |     20.9     | 54.2  |       -       |       -       |   -    |
+|    Faster R-CNN     |   X-101-32x4d-FPN   | pytorch |   1x    |     40.1     |     22.3     | 55.5  |       -       |       -       |   -    |
+|    Faster R-CNN     |   X-101-64x4d-FPN   | pytorch |   1x    |     41.3     |     23.4     | 56.6  |       -       |       -       |   -    |
+|    Faster R-CNN     |    R-50-FPN-DCN     | pytorch |   1x    |     40.0     |     22.4     | 56.1  |       -       |       -       |   -    |
+|    Faster R-CNN     | X-101-32x4d-FPN-DCN | pytorch |   1x    |     43.4     |     26.7     | 61.6  |       -       |       -       |   -    |
+|     Mask R-CNN      |      R-50-FPN       | pytorch |   1x    |     37.3     |     18.7     | 50.1  |     34.2      |     16.8      |  49.1  |
+|     Mask R-CNN      |    R-50-FPN-DCN     | pytorch |   1x    |     41.1     |     23.3     | 56.7  |     37.2      |     20.7      |  55.7  |
+|    Cascade R-CNN    |      R-50-FPN       | pytorch |   1x    |     40.4     |     20.1     | 49.7  |       -       |       -       |   -    |
+| Cascade Mask R-CNN  |      R-50-FPN       | pytorch |   1x    |     41.2     |     20.7     | 50.2  |     35.7      |     17.6      |  49.3  |
+|      RetinaNet      |      R-50-FPN       | pytorch |   1x    |     35.6     |     17.8     | 50.1  |       -       |       -       |   -    |
+| Hybrid Task Cascade | X-101-64x4d-FPN-DCN | pytorch |   1x    |     50.6     |     32.7     | 64.7  |     43.8      |     28.1      |  64.0  |
+
+Results may vary slightly due to the stochastic application of the corruptions.

docs/en/user_guides/semi_det.md

+# Semi-supervised Object Detection
+
+Semi-supervised object detection uses both labeled data and unlabeled data for training. It not only reduces the annotation burden for training high-performance object detectors but also further improves the object detector by using a large number of unlabeled data.
+
+A typical procedure to train a semi-supervised object detector is as below:
+
+- [Semi-supervised Object Detection](#semi-supervised-object-detection)
+  - [Prepare and split dataset](#prepare-and-split-dataset)
+  - [Configure multi-branch pipeline](#configure-multi-branch-pipeline)
+  - [Configure semi-supervised dataloader](#configure-semi-supervised-dataloader)
+  - [Configure semi-supervised model](#configure-semi-supervised-model)
+  - [Configure MeanTeacherHook](#configure-meanteacherhook)
+  - [Configure TeacherStudentValLoop](#configure-teacherstudentvalloop)
+
+## Prepare and split dataset
+
+We provide a dataset download script, which downloads the coco2017 dataset by default and decompresses it automatically.
+
+```shell
+python tools/misc/download_dataset.py
+```
+
+The decompressed dataset directory structure is as below:
+
+```plain
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   │   ├── image_info_unlabeled2017.json
+│   │   │   ├── instances_train2017.json
+│   │   │   ├── instances_val2017.json
+│   │   ├── test2017
+│   │   ├── train2017
+│   │   ├── unlabeled2017
+│   │   ├── val2017
+```
+
+There are two common experimental settings for semi-supervised object detection on the coco2017 dataset:
+
+(1) Split `train2017` according to a fixed percentage (1%, 2%, 5% and 10%) as a labeled dataset, and the rest of `train2017` as an unlabeled dataset. Because the different splits of `train2017` as labeled datasets will cause significant fluctuation on the accuracy of the semi-supervised detectors, five-fold cross-validation is used in practice to evaluate the algorithm. We provide the dataset split script:
+
+```shell
+python tools/misc/split_coco.py
+```
+
+By default, the script will split `train2017` according to the labeled data ratio  1%, 2%, 5% and 10%, and each split will be randomly repeated 5 times for cross-validation. The generated semi-supervised annotation file name format is as below:
+
+- the name format of labeled dataset: `instances_train2017.{fold}@{percent}.json`
+- the name format of unlabeled dataset: `instances_train2017.{fold}@{percent}-unlabeled.json`
+
+Here, `fold` is used for cross-validation, and `percent` represents the ratio of labeled data. The directory structure of the divided dataset is as below:
+
+```plain
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   │   ├── image_info_unlabeled2017.json
+│   │   │   ├── instances_train2017.json
+│   │   │   ├── instances_val2017.json
+│   │   ├── semi_anns
+│   │   │   ├── instances_train2017.1@1.json
+│   │   │   ├── instances_train2017.1@1-unlabeled.json
+│   │   │   ├── instances_train2017.1@2.json
+│   │   │   ├── instances_train2017.1@2-unlabeled.json
+│   │   │   ├── instances_train2017.1@5.json
+│   │   │   ├── instances_train2017.1@5-unlabeled.json
+│   │   │   ├── instances_train2017.1@10.json
+│   │   │   ├── instances_train2017.1@10-unlabeled.json
+│   │   │   ├── instances_train2017.2@1.json
+│   │   │   ├── instances_train2017.2@1-unlabeled.json
+│   │   ├── test2017
+│   │   ├── train2017
+│   │   ├── unlabeled2017
+│   │   ├── val2017
+```
+
+(2) Use `train2017` as the labeled dataset and `unlabeled2017` as the unlabeled dataset. Since `image_info_unlabeled2017.json` does not contain `categories` information, the `CocoDataset` cannot be initialized, so you need to write the `categories` of `instances_train2017.json` into `image_info_unlabeled2017.json` and save it as `instances_unlabeled2017.json`, the relevant script is as below:
+
+```python
+from mmengine.fileio import load, dump
+
+anns_train = load('instances_train2017.json')
+anns_unlabeled = load('image_info_unlabeled2017.json')
+anns_unlabeled['categories'] = anns_train['categories']
+dump(anns_unlabeled, 'instances_unlabeled2017.json')
+```
+
+The processed dataset directory is as below:
+
+```plain
+mmdetection
+├── data
+│   ├── coco
+│   │   ├── annotations
+│   │   │   ├── image_info_unlabeled2017.json
+│   │   │   ├── instances_train2017.json
+│   │   │   ├── instances_unlabeled2017.json
+│   │   │   ├── instances_val2017.json
+│   │   ├── test2017
+│   │   ├── train2017
+│   │   ├── unlabeled2017
+│   │   ├── val2017
+```
+
+## Configure multi-branch pipeline
+
+There are two main approaches to semi-supervised learning,
+[consistency regularization](https://research.nvidia.com/sites/default/files/publications/laine2017iclr_paper.pdf)
+and [pseudo label](https://www.researchgate.net/profile/Dong-Hyun-Lee/publication/280581078_Pseudo-Label_The_Simple_and_Efficient_Semi-Supervised_Learning_Method_for_Deep_Neural_Networks/links/55bc4ada08ae092e9660b776/Pseudo-Label-The-Simple-and-Efficient-Semi-Supervised-Learning-Method-for-Deep-Neural-Networks.pdf).
+Consistency regularization often requires some careful design, while pseudo label have a simpler form and are easier to extend to downstream tasks.
+We adopt a teacher-student joint training semi-supervised object detection framework based on pseudo label, so labeled data and unlabeled data need to configure different data pipeline:
+
+(1) Pipeline for labeled data:
+
+```python
+# pipeline used to augment labeled data,
+# which will be sent to student model for supervised training.
+sup_pipeline = [
+    dict(type='LoadImageFromFile', backend_args=backend_args),
+    dict(type='LoadAnnotations', with_bbox=True),
+    dict(type='RandomResize', scale=scale, keep_ratio=True),
+    dict(type='RandomFlip', prob=0.5),
+    dict(type='RandAugment', aug_space=color_space, aug_num=1),
+    dict(type='FilterAnnotations', min_gt_bbox_wh=(1e-2, 1e-2)),
+    dict(type='MultiBranch', sup=dict(type='PackDetInputs'))
+]
+```
+
+(2) Pipeline for unlabeled data:
+
+```python
+# pipeline used to augment unlabeled data weakly,
+# which will be sent to teacher model for predicting pseudo instances.
+weak_pipeline = [
+    dict(type='RandomResize', scale=scale, keep_ratio=True),
+    dict(type='RandomFlip', prob=0.5),
+    dict(
+        type='PackDetInputs',
+        meta_keys=('img_id', 'img_path', 'ori_shape', 'img_shape',
+                   'scale_factor', 'flip', 'flip_direction',
+                   'homography_matrix')),
+]
+
+# pipeline used to augment unlabeled data strongly,
+# which will be sent to student model for unsupervised training.
+strong_pipeline = [
+    dict(type='RandomResize', scale=scale, keep_ratio=True),
+    dict(type='RandomFlip', prob=0.5),
+    dict(
+        type='RandomOrder',
+        transforms=[
+            dict(type='RandAugment', aug_space=color_space, aug_num=1),
+            dict(type='RandAugment', aug_space=geometric, aug_num=1),
+        ]),
+    dict(type='RandomErasing', n_patches=(1, 5), ratio=(0, 0.2)),
+    dict(type='FilterAnnotations', min_gt_bbox_wh=(1e-2, 1e-2)),
+    dict(
+        type='PackDetInputs',
+        meta_keys=('img_id', 'img_path', 'ori_shape', 'img_shape',
+                   'scale_factor', 'flip', 'flip_direction',
+                   'homography_matrix')),
+]
+
+# pipeline used to augment unlabeled data into different views
+unsup_pipeline = [
+    dict(type='LoadImageFromFile', backend_args=backend_args),
+    dict(type='LoadEmptyAnnotations'),
+    dict(
+        type='MultiBranch',
+        unsup_teacher=weak_pipeline,
+        unsup_student=strong_pipeline,
+    )
+]
+```
+
+## Configure semi-supervised dataloader
+
+(1) Build a semi-supervised dataset. Use `ConcatDataset` to concatenate labeled and unlabeled datasets.
+
+```python
+labeled_dataset = dict(
+    type=dataset_type,
+    data_root=data_root,
+    ann_file='annotations/instances_train2017.json',
+    data_prefix=dict(img='train2017/'),
+    filter_cfg=dict(filter_empty_gt=True, min_size=32),
+    pipeline=sup_pipeline)
+
+unlabeled_dataset = dict(
+    type=dataset_type,
+    data_root=data_root,
+    ann_file='annotations/instances_unlabeled2017.json',
+    data_prefix=dict(img='unlabeled2017/'),
+    filter_cfg=dict(filter_empty_gt=False),
+    pipeline=unsup_pipeline)
+
+train_dataloader = dict(
+    batch_size=batch_size,
+    num_workers=num_workers,
+    persistent_workers=True,
+    sampler=dict(
+        type='GroupMultiSourceSampler',
+        batch_size=batch_size,
+        source_ratio=[1, 4]),
+    dataset=dict(
+        type='ConcatDataset', datasets=[labeled_dataset, unlabeled_dataset]))
+```
+
+(2) Use multi-source dataset sampler. Use `GroupMultiSourceSampler` to sample data form batches from `labeled_dataset` and `labeled_dataset`, `source_ratio` controls the proportion of labeled data and unlabeled data in the batch. `GroupMultiSourceSampler` also ensures that the images in the same batch have similar aspect ratios. If you don't need to guarantee the aspect ratio of the images in the batch, you can use `MultiSourceSampler`. The sampling diagram of `GroupMultiSourceSampler` is as below:
+
+<div align=center>
+<img src="https://user-images.githubusercontent.com/40661020/186149261-8cf28e92-de5c-4c8c-96e1-13558b2e27f7.jpg"/>
+</div>
+
+`sup=1000` indicates that the scale of the labeled dataset is 1000, `sup_h=200` indicates that the scale of the images with an aspect ratio greater than or equal to 1 in the labeled dataset is 200, and `sup_w=800` indicates that the scale of the images with an aspect ratio less than 1 in the labeled dataset is 800,
+`unsup=9000` indicates that the scale of the unlabeled dataset is 9000, `unsup_h=1800` indicates that the scale of the images with an aspect ratio greater than or equal to 1 in the unlabeled dataset is 1800, and `unsup_w=7200` indicates the scale of the images with an aspect ratio less than 1 in the unlabeled dataset is 7200.
+`GroupMultiSourceSampler` randomly selects a group according to the overall aspect ratio distribution of the images in the labeled dataset and the unlabeled dataset, and then sample data to form batches from the two datasets according to `source_ratio`, so labeled datasets and unlabeled datasets have different repetitions.
+
+## Configure semi-supervised model
+
+We choose `Faster R-CNN` as `detector` for semi-supervised training. Take the semi-supervised object detection algorithm `SoftTeacher` as an example,
+the model configuration can be inherited from `_base_/models/faster-rcnn_r50_fpn.py`, replacing the backbone network of the detector with `caffe` style.
+Note that unlike the supervised training configs, `Faster R-CNN` as `detector` is an attribute of `model`, not `model` .
+In addition, `data_preprocessor` needs to be set to `MultiBranchDataPreprocessor`, which is used to pad and normalize images from different pipelines.
+Finally, parameters required for semi-supervised training and testing can be configured via `semi_train_cfg` and `semi_test_cfg`.
+
+```python
+_base_ = [
+    '../_base_/models/faster-rcnn_r50_fpn.py', '../_base_/default_runtime.py',
+    '../_base_/datasets/semi_coco_detection.py'
+]
+
+detector = _base_.model
+detector.data_preprocessor = dict(
+    type='DetDataPreprocessor',
+    mean=[103.530, 116.280, 123.675],
+    std=[1.0, 1.0, 1.0],
+    bgr_to_rgb=False,
+    pad_size_divisor=32)
+detector.backbone = dict(
+    type='ResNet',
+    depth=50,
+    num_stages=4,
+    out_indices=(0, 1, 2, 3),
+    frozen_stages=1,
+    norm_cfg=dict(type='BN', requires_grad=False),
+    norm_eval=True,
+    style='caffe',
+    init_cfg=dict(
+        type='Pretrained',
+        checkpoint='open-mmlab://detectron2/resnet50_caffe'))
+
+model = dict(
+    _delete_=True,
+    type='SoftTeacher',
+    detector=detector,
+    data_preprocessor=dict(
+        type='MultiBranchDataPreprocessor',
+        data_preprocessor=detector.data_preprocessor),
+    semi_train_cfg=dict(
+        freeze_teacher=True,
+        sup_weight=1.0,
+        unsup_weight=4.0,
+        pseudo_label_initial_score_thr=0.5,
+        rpn_pseudo_thr=0.9,
+        cls_pseudo_thr=0.9,
+        reg_pseudo_thr=0.02,
+        jitter_times=10,
+        jitter_scale=0.06,
+        min_pseudo_bbox_wh=(1e-2, 1e-2)),
+    semi_test_cfg=dict(predict_on='teacher'))
+```
+
+In addition, we also support semi-supervised training for other detection models, such as `RetinaNet` and `Cascade R-CNN`. Since `SoftTeacher` only supports `Faster R-CNN`, it needs to be replaced with `SemiBaseDetector`, example is as below:
+
+```python
+_base_ = [
+    '../_base_/models/retinanet_r50_fpn.py', '../_base_/default_runtime.py',
+    '../_base_/datasets/semi_coco_detection.py'
+]
+
+detector = _base_.model
+
+model = dict(
+    _delete_=True,
+    type='SemiBaseDetector',
+    detector=detector,
+    data_preprocessor=dict(
+        type='MultiBranchDataPreprocessor',
+        data_preprocessor=detector.data_preprocessor),
+    semi_train_cfg=dict(
+        freeze_teacher=True,
+        sup_weight=1.0,
+        unsup_weight=1.0,
+        cls_pseudo_thr=0.9,
+        min_pseudo_bbox_wh=(1e-2, 1e-2)),
+    semi_test_cfg=dict(predict_on='teacher'))
+```
+
+Following the semi-supervised training configuration of `SoftTeacher`, change `batch_size` to 2 and `source_ratio` to `[1, 1]`, the experimental results of supervised and semi-supervised training of `RetinaNet`, `Faster R-CNN`, `Cascade R-CNN` and `SoftTeacher` on the 10% coco `train2017` are as below:
+
+|      Model       |   Detector    | BackBone | Style | sup-0.1-coco mAP | semi-0.1-coco mAP |
+| :--------------: | :-----------: | :------: | :---: | :--------------: | :---------------: |
+| SemiBaseDetector |   RetinaNet   | R-50-FPN | caffe |       23.5       |       27.7        |
+| SemiBaseDetector | Faster R-CNN  | R-50-FPN | caffe |       26.7       |       28.4        |
+| SemiBaseDetector | Cascade R-CNN | R-50-FPN | caffe |       28.0       |       29.7        |
+|   SoftTeacher    | Faster R-CNN  | R-50-FPN | caffe |       26.7       |       31.1        |
+
+## Configure MeanTeacherHook
+
+Usually, the teacher model is updated by Exponential Moving Average (EMA) the student model, and then the teacher model is optimized with the optimization of the student model, which can be achieved by configuring `custom_hooks`:
+
+```python
+custom_hooks = [dict(type='MeanTeacherHook')]
+```
+
+## Configure TeacherStudentValLoop
+
+Since there are two models in the teacher-student joint training framework, we can replace `ValLoop` with `TeacherStudentValLoop` to test the accuracy of both models during the training process.
+
+```python
+val_cfg = dict(type='TeacherStudentValLoop')
+```

docs/en/user_guides/single_stage_as_rpn.md

+# Use a single stage detector as RPN
+
+Region proposal network (RPN) is a submodule in [Faster R-CNN](https://arxiv.org/abs/1506.01497), which generates proposals for the second stage of Faster R-CNN. Most two-stage detectors in MMDetection use [`RPNHead`](../../../mmdet/models/dense_heads/rpn_head.py) to generate proposals as RPN. However, any single-stage detector can serve as an RPN since their bounding box predictions can also be regarded as region proposals and thus be refined in the R-CNN. Therefore, MMDetection v3.0 supports that.
+
+To illustrate the whole process, here we give an example of how to use an anchor-free single-stage model [FCOS](../../../configs/fcos/fcos_r50-caffe_fpn_gn-head_1x_coco.py) as an RPN in [Faster R-CNN](../../../configs/faster_rcnn/faster-rcnn_r50_fpn_fcos-rpn_1x_coco.py).
+
+The outline of this tutorial is as below:
+
+1. Use `FCOSHead` as an `RPNHead` in Faster R-CNN
+2. Evaluate proposals
+3. Train the customized Faster R-CNN with pre-trained FCOS
+
+## Use `FCOSHead` as an `RPNHead` in Faster R-CNN
+
+To set `FCOSHead` as an `RPNHead` in Faster R-CNN, we should create a new config file named `configs/faster_rcnn/faster-rcnn_r50_fpn_fcos-rpn_1x_coco.py`, and replace with the setting of `rpn_head` with the setting of `bbox_head` in `configs/fcos/fcos_r50-caffe_fpn_gn-head_1x_coco.py`. Besides, we still use the neck setting of FCOS with strides of `[8, 16, 32, 64, 128]`, and update `featmap_strides` of `bbox_roi_extractor` to `[8, 16, 32, 64, 128]`. To avoid loss goes NAN, we apply warmup during the first 1000 iterations instead of the first 500 iterations, which means that the lr increases more slowly. The config is as follows:
+
+```python
+_base_ = [
+    '../_base_/models/faster-rcnn_r50_fpn.py',
+    '../_base_/datasets/coco_detection.py',
+    '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py'
+]
+
+model = dict(
+    # copied from configs/fcos/fcos_r50-caffe_fpn_gn-head_1x_coco.py
+    neck=dict(
+        start_level=1,
+        add_extra_convs='on_output',  # use P5
+        relu_before_extra_convs=True),
+    rpn_head=dict(
+        _delete_=True,  # ignore the unused old settings
+        type='FCOSHead',
+        num_classes=1,  # num_classes = 1 for rpn, if num_classes > 1, it will be set to 1 in TwoStageDetector automatically
+        in_channels=256,
+        stacked_convs=4,
+        feat_channels=256,
+        strides=[8, 16, 32, 64, 128],
+        loss_cls=dict(
+            type='FocalLoss',
+            use_sigmoid=True,
+            gamma=2.0,
+            alpha=0.25,
+            loss_weight=1.0),
+        loss_bbox=dict(type='IoULoss', loss_weight=1.0),
+        loss_centerness=dict(
+            type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0)),
+    roi_head=dict(  # update featmap_strides due to the strides in neck
+        bbox_roi_extractor=dict(featmap_strides=[8, 16, 32, 64, 128])))
+
+# learning rate
+param_scheduler = [
+    dict(
+        type='LinearLR', start_factor=0.001, by_epoch=False, begin=0,
+        end=1000),  # Slowly increase lr, otherwise loss becomes NAN
+    dict(
+        type='MultiStepLR',
+        begin=0,
+        end=12,
+        by_epoch=True,
+        milestones=[8, 11],
+        gamma=0.1)
+]
+```
+
+Then, we could use the following command to train our customized model. For more training commands, please refer to [here](train.md).
+
+```python
+# training with 8 GPUS
+bash tools/dist_train.sh configs/faster_rcnn/faster-rcnn_r50_fpn_fcos-rpn_1x_coco.py \
+    8 \
+    --work-dir ./work_dirs/faster-rcnn_r50_fpn_fcos-rpn_1x_coco
+```
+
+## Evaluate proposals
+
+The quality of proposals is of great importance to the performance of detector, therefore, we also provide a way to evaluate proposals. Same as above, create a new config file named `configs/rpn/fcos-rpn_r50_fpn_1x_coco.py`, and replace with setting of `rpn_head` with the setting of `bbox_head` in `configs/fcos/fcos_r50-caffe_fpn_gn-head_1x_coco.py`.
+
+```python
+_base_ = [
+    '../_base_/models/rpn_r50_fpn.py', '../_base_/datasets/coco_detection.py',
+    '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py'
+]
+
+val_evaluator = dict(metric='proposal_fast')
+test_evaluator = val_evaluator
+
+model = dict(
+    # copied from configs/fcos/fcos_r50-caffe_fpn_gn-head_1x_coco.py
+    neck=dict(
+        start_level=1,
+        add_extra_convs='on_output',  # use P5
+        relu_before_extra_convs=True),
+    rpn_head=dict(
+        _delete_=True,  # ignore the unused old settings
+        type='FCOSHead',
+        num_classes=1,  # num_classes = 1 for rpn, if num_classes > 1, it will be set to 1 in RPN automatically
+        in_channels=256,
+        stacked_convs=4,
+        feat_channels=256,
+        strides=[8, 16, 32, 64, 128],
+        loss_cls=dict(
+            type='FocalLoss',
+            use_sigmoid=True,
+            gamma=2.0,
+            alpha=0.25,
+            loss_weight=1.0),
+        loss_bbox=dict(type='IoULoss', loss_weight=1.0),
+        loss_centerness=dict(
+            type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0)))
+```
+
+Suppose we have the checkpoint `./work_dirs/faster-rcnn_r50_fpn_fcos-rpn_1x_coco/epoch_12.pth` after training, then we can evaluate the quality of proposals with the following command.
+
+```python
+# testing with 8 GPUs
+bash tools/dist_test.sh \
+    configs/rpn/fcos-rpn_r50_fpn_1x_coco.py \
+    ./work_dirs/faster-rcnn_r50_fpn_fcos-rpn_1x_coco/epoch_12.pth \
+    8
+```
+
+## Train the customized Faster R-CNN with pre-trained FCOS
+
+Pre-training not only speeds up convergence of training, but also improves the performance of the detector. Therefore, here we give an example to illustrate how to do use a pre-trained FCOS as an RPN to accelerate training and improve the accuracy. Suppose we want to use `FCOSHead` as an rpn head in Faster R-CNN and train with the pre-trained [`fcos_r50-caffe_fpn_gn-head_1x_coco`](https://download.openmmlab.com/mmdetection/v2.0/fcos/fcos_r50_caffe_fpn_gn-head_1x_coco/fcos_r50_caffe_fpn_gn-head_1x_coco-821213aa.pth). The content of config file named `configs/faster_rcnn/faster-rcnn_r50-caffe_fpn_fcos-rpn_1x_coco.py` is as the following. Note that `fcos_r50-caffe_fpn_gn-head_1x_coco` uses a caffe version of ResNet50, the pixel mean and std in `data_preprocessor` thus need to be updated.
+
+```python
+_base_ = [
+    '../_base_/models/faster-rcnn_r50_fpn.py',
+    '../_base_/datasets/coco_detection.py',
+    '../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py'
+]
+
+model = dict(
+    data_preprocessor=dict(
+        mean=[103.530, 116.280, 123.675],
+        std=[1.0, 1.0, 1.0],
+        bgr_to_rgb=False),
+    backbone=dict(
+        norm_cfg=dict(type='BN', requires_grad=False),
+        style='caffe',
+        init_cfg=None),  # the checkpoint in ``load_from`` contains the weights of backbone
+    neck=dict(
+        start_level=1,
+        add_extra_convs='on_output',  # use P5
+        relu_before_extra_convs=True),
+    rpn_head=dict(
+        _delete_=True,  # ignore the unused old settings
+        type='FCOSHead',
+        num_classes=1,  # num_classes = 1 for rpn, if num_classes > 1, it will be set to 1 in TwoStageDetector automatically
+        in_channels=256,
+        stacked_convs=4,
+        feat_channels=256,
+        strides=[8, 16, 32, 64, 128],
+        loss_cls=dict(
+            type='FocalLoss',
+            use_sigmoid=True,
+            gamma=2.0,
+            alpha=0.25,
+            loss_weight=1.0),
+        loss_bbox=dict(type='IoULoss', loss_weight=1.0),
+        loss_centerness=dict(
+            type='CrossEntropyLoss', use_sigmoid=True, loss_weight=1.0)),
+    roi_head=dict(  # update featmap_strides due to the strides in neck
+        bbox_roi_extractor=dict(featmap_strides=[8, 16, 32, 64, 128])))
+
+load_from = 'https://download.openmmlab.com/mmdetection/v2.0/fcos/fcos_r50_caffe_fpn_gn-head_1x_coco/fcos_r50_caffe_fpn_gn-head_1x_coco-821213aa.pth'
+```
+
+The command for training is as below.
+
+```python
+bash tools/dist_train.sh \
+    configs/faster_rcnn/faster-rcnn_r50-caffe_fpn_fcos-rpn_1x_coco.py \
+    8 \
+    --work-dir ./work_dirs/faster-rcnn_r50-caffe_fpn_fcos-rpn_1x_coco
+```