changelog.md

Changelog of v0.x

v0.10.5 (11/9/2024)

• Fix _is_builtin_module. by @HAOCHENYE in open-mmlab#1571

v0.10.4 (23/4/2024)

New Features & Enhancements

• Support custom artifact_location in MLflowVisBackend. by @daavoo in open-mmlab#1505
• Add the supported pytorch versions in README by @zhouzaida in open-mmlab#1512
• Perform evaluation upon training completion by @LZHgrla in open-mmlab#1529
• Enable exclude_frozen_parameters for DeepSpeedEngine._zero3_consolidated_16bit_state_dict by @LZHgrla in open-mmlab#1517

Bug Fixes

• Fix warning capture by @fanqiNO1 in open-mmlab#1494
• Remove codeowners file by @zhouzaida in open-mmlab#1496
• Fix config of readthedocs by @zhouzaida in open-mmlab#1511
• Delete frozen parameters when using paramwise_cfg by @LZHgrla in open-mmlab#1441

Docs

• Refine mmengine intro by @zhouzaida in open-mmlab#1479
• Fix typo by @zhouzaida in open-mmlab#1481
• Fix typos and remove fullwidth unicode chars by @evdcush in open-mmlab#1488
• Fix docstring of Config by @MambaWong in open-mmlab#1506
• Fix typo by @hiramf in open-mmlab#1532

v0.10.3 (24/1/2024)

New Features & Enhancements

• Add the support for musa device support by @hanhaowen-mt in open-mmlab#1453
• Support save_optimizer=False for DeepSpeed by @LZHgrla in open-mmlab#1474
• Update visualizer.py by @Anm-pinellia in open-mmlab#1476

Bug Fixes

• Fix Config.to_dict by @HAOCHENYE in open-mmlab#1465
• Fix the resume of iteration by @LZHgrla in open-mmlab#1471
• Fix dist.collect_results to keep all ranks' elements by @LZHgrla in open-mmlab#1469

Docs

• Add the usage of ProfilerHook by @zhouzaida in open-mmlab#1466
• Fix the nnodes in the doc of ddp training by @XiwuChen in open-mmlab#1462

v0.10.2 (26/12/2023)

New Features & Enhancements

• Support multi-node distributed training with NPU backend by @shun001 in open-mmlab#1459
• Use ImportError to cover ModuleNotFoundError by @del-zhenwu in open-mmlab#1438

Bug Fixes

• Fix bug in load_model_state_dict of BaseStrategy by @SCZwangxiao in open-mmlab#1447
• Fix placement policy in ColossalAIStrategy by @fanqiNO1 in open-mmlab#1440

Contributors

A total of 4 developers contributed to this release. Thanks @shun001, @del-zhenwu, @SCZwangxiao, @fanqiNO1

v0.10.1 (22/11/2023)

Bug Fixes

• Fix collect_env without opencv by @fanqiNO1 in open-mmlab#1434
• Fix deploy.yml by @fanqiNO1 in open-mmlab#1431

Docs

• Add build mmengine-lite from source by @fanqiNO1 in open-mmlab#1435

Contributors

A total of 1 developers contributed to this release. Thanks @fanqiNO1

v0.10.0 (21/11/2023)

New Features & Enhancements

• Support for installing mmengine without opencv by @fanqiNO1 in open-mmlab#1429
• Support exclude_frozen_parameters for DeepSpeedStrategy's resume by @LZHgrla in open-mmlab#1424

Bug Fixes

• Fix bugs in colo optimwrapper by @HIT-cwh in open-mmlab#1426
• Fix scale_lr in SingleDeviceStrategy by @fanqiNO1 in open-mmlab#1428
• Fix CI for torch2.1.0 by @fanqiNO1 in open-mmlab#1418

Contributors

A total of 3 developers contributed to this release. Thanks @HIT-cwh, @LZHgrla, @fanqiNO1

v0.9.1 (03/11/2023)

New Features & Enhancements

• Support slurm distributed training for mlu devices by @POI-WX in open-mmlab#1396
• Add torch 2.1.0 checking in CI by @YiyaoYang1 in open-mmlab#1389
• Add exclude_frozen_parameters for DeepSpeedStrategy by @LZHgrla in open-mmlab#1415
• Enhance inputs_to_half in DeepSpeedStrategy by @fanqiNO1 in open-mmlab#1400

Bug Fixes

• Fix new config in visualizer by @HAOCHENYE in open-mmlab#1390
• Fix func params using without init in OneCycleLR (#1401) by @whlook in open-mmlab#1403
• Fix a bug when module is missing in low version of bitsandbytes by @Ben-Louis in open-mmlab#1388
• Fix ConcatDataset raising error when metainfo is np.array by @jonbakerfish in open-mmlab#1407

Docs

• Rename master to main by @zhouzaida in open-mmlab#1397

Contributors

A total of 9 developers contributed to this release. Thanks @POI-WX, @whlook, @jonbakerfish, @LZHgrla, @Ben-Louis, @YiyaoYang1, @fanqiNO1, @HAOCHENYE, @zhouzaida

v0.9.0 (10/10/2023)

Highlights

• Support training with ColossalAI. Refer to the Training Large Models for more detailed usages.
• Support gradient checkpointing. Refer to the Save Memory on GPU for more details.
• Supports multiple visualization backends, including NeptuneVisBackend, DVCLiveVisBackend and AimVisBackend. Refer to Visualization Backends for more details.

New Features & Enhancements

• Add a text translation example by @Desjajja in open-mmlab#1283
• Add NeptuneVisBackend by @wangerlie in open-mmlab#1311
• Add ColossalAI strategy by @HAOCHENYE in open-mmlab#1299
• Add collect_results support for Ascend NPU by @xuuyangg in open-mmlab#1309
• Unify the parameter style of DeepSpeedStrategy by @LZHgrla in open-mmlab#1320
• Add progressbar rich by @Dominic23331 in open-mmlab#1157
• Support using other file handlers by @KevinNuNu in open-mmlab#1188
• Refine error message by @zhouzaida in open-mmlab#1338
• Implement gradient checkpointing by @zeyuanyin in open-mmlab#1319
• Add DVCLiveVisBackend by @RangeKing in open-mmlab#1336
• Add AimVisBackend by @RangeKing in open-mmlab#1347
• Support bitsandbytes by @okotaku in open-mmlab#1357
• Support Adafactor Optimizer by @okotaku in open-mmlab#1361
• Add unit tests for autocast with Ascend device by @6Vvv in open-mmlab#1363
• Support metainfo of dataset can be a generic dict-like Mapping by @hiyyg in open-mmlab#1378
• Support for installing minimal runtime dependencies by @HAOCHENYE in open-mmlab#1362
• Support setting the number of iterations in Runner for each epoch by @ShuRaymond in open-mmlab#1292
• Support using gradient checkpointing in FSDP by @HAOCHENYE in open-mmlab#1382

Docs

• Add README for examples by @zhouzaida in open-mmlab#1295
• Add a new ecosystem in README by @zhouzaida in open-mmlab#1296
• Fix typo by @zhouzaida in open-mmlab#1298
• Add an image for Neptune by @zhouzaida in open-mmlab#1312
• Fix docs of ColossalAI by @HAOCHENYE in open-mmlab#1315
• Update QRCode by @crazysteeaam in open-mmlab#1328
• Add activation checkpointing usage by @zhouzaida in open-mmlab#1341
• Fix typo by @zhouzaida in open-mmlab#1348
• Update the usage of bitsandbytes in Chinese documents by @zhouzaida in open-mmlab#1359
• Fix doc typo our_dir in LoggerHook by @wangg12 in open-mmlab#1373
• Add the contributing doc in pr template by @zhouzaida in open-mmlab#1380
• Update the version info by @fanqiNO1 in open-mmlab#1383
• Fix typo by @fanqiNO1 in open-mmlab#1385

Bug Fixes

• Ignore examples in CI by @zhouzaida in open-mmlab#1297
• Fix multi-card issue in PyTorch v2.1 on Ascend by @LRJKD in open-mmlab#1321
• Fix get optimizer_cls by @HAOCHENYE in open-mmlab#1324
• Fix ndarray metainfo check in ConcatDataset by @NrealLzx in open-mmlab#1333
• Adapt to PyTorch v2.1 on Ascend by @LRJKD in open-mmlab#1332
• Fix the type check of tasks in progress bar by @zhouzaida in open-mmlab#1340
• The keyword mode appears nested multiple times in the log by @huaibovip in open-mmlab#1305
• Fix pydantic version to fix mlflow unit tests by @zhouzaida in open-mmlab#1351
• Fix get class attribute from a string by @HAOCHENYE in open-mmlab#1345
• Delete yapf verify by @okotaku in open-mmlab#1365
• Ensure from_cfg of Runner have the same defaults values as its init by @YinAoXiong in open-mmlab#1368
• Fix docs building error caused by deepspeed by @HAOCHENYE in open-mmlab#1379

Contributors

A total of 21 developers contributed to this release. Thanks @LZHgrla, @wangerlie, @wangg12, @RangeKing, @hiyyg, @LRJKD, @KevinNuNu, @zeyuanyin, @Desjajja, @ShuRaymond, @okotaku, @crazysteeaam, @6Vvv, @NrealLzx, @YinAoXiong, @huaibovip, @xuuyangg, @Dominic23331, @fanqiNO1, @HAOCHENYE, @zhouzaida

v0.8.4 (03/08/2023)

New Features & Enhancements

• Support callable collate_fn for FlexibleRunner by @LZHgrla in open-mmlab#1284

Bug fixes

• Skip adding vis_backends when save_dir is not set by @HAOCHENYE in open-mmlab#1289
• Fix dumping pure python style config in colab by @HAOCHENYE in open-mmlab#1290

Docs

• Find unused parameters by @zhouzaida in open-mmlab#1288

Contributors

A total of 3 developers contributed to this release. Thanks @HAOCHENYE, @zhouzaida, @LZHgrla

v0.8.3 (31/07/2023)

Highlights

• Support enabling efficient_conv_bn_eval for efficient convolution and batch normalization. See save memory on gpu for more details
• Add Llama2 finetune example
• Support multi-node distributed training with MLU backend

New Features & Enhancements

• Enable efficient_conv_bn_eval for memory saving convolution and batch normalization by @youkaichao in open-mmlab#1202, open-mmlab#1251 and open-mmlab#1259
• Add Llama2 example by @HAOCHENYE in open-mmlab#1264
• Compare the difference of two configs by @gachiemchiep in open-mmlab#1260
• Enable explicit error for deepspeed not installed by @Li-Qingyun in open-mmlab#1240
• Support skipping initialization in BaseModule by @HAOCHENYE in open-mmlab#1263
• Add parameter save_begin to control when to save checkpoints by @KerwinKai in open-mmlab#1271
• Support multi-node distributed training with MLU backend by @josh6688 in open-mmlab#1266
• Enhance error message thrown by Config, build function and ConfigDict.items by @HAOCHENYE in open-mmlab#1272, open-mmlab#1270 and open-mmlab#1088
• Add the loop_stage runtime information in message_hub by @zhouzaida in open-mmlab#1277
• Fix Visualizer that built vis_backends will not be used when save_dir is None by @Xinyu302 in open-mmlab#1275

Bug fixes

• Fix scalar check in RuntimeInfoHook by @i-aki-y in open-mmlab#1250
• Move data preprocessor to target device in FSDPStrategy by @HAOCHENYE in open-mmlab#1261

Docs

• Add ecosystem in README by @zhouzaida in open-mmlab#1247
• Add short explanation about registry scope by @mmeendez8 in open-mmlab#1114
• Add the data flow of Runner in README by @zhouzaida in open-mmlab#1257
• Introduce how to customize distributed training settings by @zhouzaida in open-mmlab#1279

Contributors

A total of 9 developers contributed to this release. Thanks @HAOCHENYE, @youkaichao, @josh6688, @i-aki-y, @mmeendez8, @zhouzaida, @gachiemchiep, @KerwinKai, @Li-Qingyun

v0.8.2 (07/12/2023)

Bug fixes

• Fix pickling the Python style config by @HAOCHENYE in open-mmlab#1241
• Fix the logic of setting lazy_import by @Li-Qingyun in open-mmlab#1239

v0.8.1 (07/05/2023)

New Features & Enhancements

• Accelerate Config.dump and support converting Lazyxxx to string in ConfigDict.to_dictby @HAOCHENYE in open-mmlab#1232

Bug fixes

• FSDP should call _get_ignored_modules by @HAOCHENYE in open-mmlab#1235

Docs

• Add a document to introduce how to train a large model by @zhouzaida in open-mmlab#1228

v0.8.0 (06/30/2023)

Highlights

• Support training with FSDP and DeepSpeed. Refer to the example for more detailed usages.

• Introduce the pure Python style configuration file:

  • Support navigating to base configuration file in IDE
  • Support navigating to base variable in IDE
  • Support navigating to source code of class in IDE
  • Support inheriting two configuration files containing the same field
  • Load the configuration file without other third-party requirements

  Refer to the tutorial for more detailed usages.

  

New Features & Enhancements

• Support training with FSDP by @HAOCHENYE in open-mmlab#1213
• Add FlexibleRunner and Strategies, and support training with DeepSpeed by @zhouzaida in open-mmlab#1183
• Support pure Python style configuration file by @HAOCHENYE in open-mmlab#1071
• Learning rate in log can show the base learning rate of optimizer by @AkideLiu in open-mmlab#1019
• Refine the error message when auto_scale_lr is not set correctly by @alexander-soare in open-mmlab#1181
• WandbVisBackend supports updating config by @zgzhengSEU in open-mmlab#977

Bug fixes

• CheckpointHook should check whether file exists before removing it by @zhouzaida in open-mmlab#1198
• Fix undefined variable error in Runner by @HAOCHENYE in open-mmlab#1219

Docs

• Add a document to introduce how to debug with vscode by @zhouzaida in open-mmlab#1212
• Update English introduction by @evdcush in open-mmlab#1189
• Fix parameter typing error in document by @syo093c in open-mmlab#1201
• Fix gpu collection during evaluation by @edkair in open-mmlab#1208
• Fix a comment in runner tutorial by @joihn in open-mmlab#1210

Contributors

A total of 9 developers contributed to this release. Thanks @evdcush, @zhouzaida, @AkideLiu, @joihn, @HAOCHENYE, @edkair, @alexander-soare, @syo093c, @zgzhengSEU

v0.7.4 (06/03/2023)

Highlights

• Support using ClearML to record experiment data
• Add Sophia optimizers

New Features & Enhancements

• Add visualize backend for clearml by @gachiemchiep in open-mmlab#1091
• Support Sophia optimizers by @zhouzaida in open-mmlab#1170
• Refactor unittest syncbuffer by @HAOCHENYE in open-mmlab#813
• Allow ann_file, data_root is None for BaseDataset by @HAOCHENYE in open-mmlab#850
• Enable full precision training on Ascend NPU by @Ginray in open-mmlab#1109
• Creating a text classification example by @TankNee in open-mmlab#1122
• Add option to log selected config only by @KickCellarDoor in open-mmlab#1159
• Add an option to control whether to show progress bar in BaseInference by @W-ZN in open-mmlab#1135
• Support dipu device by @CokeDong in open-mmlab#1127
• Let unit tests not affect each other by @zhouzaida in open-mmlab#1169
• Add support for full wandb's define_metric arguments by @i-aki-y in open-mmlab#1099

Bug fixes

• Fix the incorrect device of inputs in get_model_complexity_info by @CescMessi in open-mmlab#1130
• Correctly saves _metadata of state_dict when saving checkpoints by @Bomsw in open-mmlab#1131
• Correctly record random seed in log by @Shiyang980713 in open-mmlab#1152
• Close MLflowVisBackend only if active by @zimonitrome in open-mmlab#1151
• Fix ProfileHook cannot profile ddp-training by @HAOCHENYE in open-mmlab#1140
• Handle the case for Multi-Instance GPUs when using cuda_visible_devices by @adrianjoshua-strutt in open-mmlab#1164
• Fix attribute error when parsing CUDA_VISIBLE_DEVICES in logger @Xiangxu-0103 in open-mmlab#1172

Docs

• Translate infer.md by @Hongru-Xiao in open-mmlab#1121
• Fix a missing comma in tutorials/runner.md by @gy-7 in open-mmlab#1146
• Fix typo in comment by @YQisme in open-mmlab#1154
• Translate data_element.md by @xin-li-67 in open-mmlab#1067
• Add the usage of clearml by @zhouzaida in open-mmlab#1180

Contributors

A total of 19 developers contributed to this release. Thanks @Hongru-Xiao @i-aki-y @Bomsw @KickCellarDoor @zhouzaida @YQisme @gachiemchiep @CescMessi @W-ZN @Ginray @adrianjoshua-strutt @CokeDong @xin-li-67 @Xiangxu-0103 @HAOCHENYE @Shiyang980713 @TankNee @zimonitrome @gy-7

v0.7.3 (04/28/2023)

Highlights

• Support using MLflow to record experiment data
• Support registering callable objects to the registry

New Features & Enhancements

• Add MLflowVisBackend by @sh0622-kim in open-mmlab#878
• Support customizing worker_init_fn in dataloader config by @shufanwu in open-mmlab#1038
• Make the parameters of get_model_complexity_info() friendly by @sjiang95 in open-mmlab#1056
• Add torch_npu optimizer by @luomaoling in open-mmlab#1079
• Support registering callable objects @C1rN09 in open-mmlab#595
• Complement type hint of get_model_complexity_info() by @sjiang95 in open-mmlab#1064
• MessageHub.get_info() supports returning a default value by @enkilee in open-mmlab#991
• Refactor logger hook unit test by @HAOCHENYE in open-mmlab#797
• Support BoolTensor and LongTensor on Ascend NPU by @Ginray in open-mmlab#1011
• Remove useless variable declaration by @HAOCHENYE in open-mmlab#1052
• Enhance the support for MLU device by @josh6688 in open-mmlab#1075
• Support configuring synchronization directory for BaseMetric by @HAOCHENYE in open-mmlab#1074
• Support accepting multiple input_shape for get_model_complexity_info by @sjiang95 in open-mmlab#1065
• Enhance docstring and error catching in MessageHub by @HAOCHENYE in open-mmlab#1098
• Enhance the efficiency of Visualizer.show by @HAOCHENYE in open-mmlab#1015
• Update repo list by @HAOCHENYE in open-mmlab#1108
• Enhance error message during custom import by @HAOCHENYE in open-mmlab#1102
• Support _load_state_dict_post_hooks in load_state_dict by @mzr1996 in open-mmlab#1103

Bug fixes

• Fix publishing multiple checkpoints when using multiple GPUs by @JunweiZheng93 in open-mmlab#1070
• Fix error when log_with_hierarchy is True by @HAOCHENYE in open-mmlab#1085
• Call SyncBufferHook before validation in IterBasedTrainLoop by @Luo-Yihang in open-mmlab#982
• Fix the resuming error caused by HistoryBuffer by @HAOCHENYE in open-mmlab#1078
• Failed to remove the previous best checkpoints by @HAOCHENYE in open-mmlab#1086
• Fix using incorrect local rank by @C1rN09 in open-mmlab#973
• No training log when the num of iterations is smaller than the default interval by @shufanwu in open-mmlab#1046
• collate_fn could not be a function object by @zhouzaida in open-mmlab#1093
• Fix optimizer.state could be saved in cuda:0 by @HAOCHENYE in open-mmlab#966
• Fix building unnecessary loop during train/test/val by @HAOCHENYE in open-mmlab#1107

Docs

• Introduce the use of wandb and tensorboard by @zhouzaida in open-mmlab#912
• Translate tutorials/evaluation.md by @LEFTeyex in open-mmlab#1053
• Translate design/evaluation.md by @zccjjj in open-mmlab#1062
• Fix three typos in runner by @jsrdcht in open-mmlab#1068
• Translate migration/hook.md to English by @SheffieldCao in open-mmlab#1054
• Replace MMCls with MMPretrain in docs by @zhouzaida in open-mmlab#1096

Contributors

A total of 17 developers contributed to this release. Thanks @enkilee, @JunweiZheng93, @sh0622-kim, @jsrdcht, @SheffieldCao, @josh6688, @mzr1996, @zhouzaida, @shufanwu, @Luo-Yihang, @C1rN09, @LEFTeyex, @zccjjj, @Ginray, @HAOCHENYE, @sjiang95, @luomaoling

v0.7.2 (04/06/2023)

Bug fixes

• Align the evaluation result in log by @kitecats in open-mmlab#1034
• Update the logic to calculate the repeat_factors in ClassBalancedDataset by @BIGWangYuDong in open-mmlab#1048
• Initialize sub-modules in DistributedDataParallel that define init_weights during initialization by @HAOCHENYE in open-mmlab#1045
• Refactor checkpointhook unittest by @HAOCHENYE in open-mmlab#789

Contributors

A total of 3 developers contributed to this release. Thanks @kitecats, @BIGWangYuDong, @HAOCHENYE

v0.7.1 (04/03/2023)

Highlights

• Support compiling the model and enabling mixed-precision training at the same time
• Fix the bug where the logs cannot be properly saved to the log file after calling torch.compile

New Features & Enhancements

• Add mmpretrain to the MODULE2PACKAGE. by @mzr1996 in open-mmlab#1002
• Support using get_device in the compiled model by @C1rN09 in open-mmlab#1004
• Make sure the FileHandler still alive after torch.compile by @HAOCHENYE in open-mmlab#1021
• Unify the use of print_log and logger.info(warning) by @LEFTeyex in open-mmlab#997
• Publish models after training if published_keys is set in CheckpointHook by @KerwinKai in open-mmlab#987
• Enhance the error catching in registry by @HAOCHENYE in open-mmlab#1010
• Do not print config if it is empty by @zhouzaida in open-mmlab#1028

Bug fixes

• Fix there is no space between data_time and metric in logs by @HAOCHENYE in open-mmlab#1025

Docs

• Minor fixes in EN docs to remove or replace unicode chars with ascii by @evdcush in open-mmlab#1018

Contributors

A total of 7 developers contributed to this release. Thanks @LEFTeyex, @KerwinKai, @mzr1996, @evdcush, @C1rN09, @HAOCHENYE, @zhouzaida

v0.7.0 (03/16/2023)

Highlights

• Support PyTorch 2.0! Accelerate training by compiling models. See the tutorial Model Compilation for details
• Add EarlyStoppingHook to stop training when the metric does not improve

New Features & Enhancements

• Add configurations to support torch.compile in Runner by @C1rN09 in open-mmlab#976
• Support EarlyStoppingHook by @nijkah in open-mmlab#739
• Disable duplicated warning during distributed training by @HAOCHENYE in open-mmlab#961
• Add FUNCTIONS root Registry by @HAOCHENYE in open-mmlab#983
• Save the "memory" field to visualization backends by @enkilee in open-mmlab#974
• Enable bf16 in AmpOptimWrapper by @C1rN09 in open-mmlab#960
• Support writing data to vis_backend with prefix by @HAOCHENYE in open-mmlab#972
• Support exporting logs of different ranks in debug mode by @HAOCHENYE in open-mmlab#968
• Silence error when ManagerMixin built instance with duplicate name. by @HAOCHENYE in open-mmlab#990

Bug fixes

• Fix optim_wrapper unittest for pytorch < 1.10.0 by @C1rN09 in open-mmlab#975
• Support calculating the flops of matmul with single dimension matrix by @HAOCHENYE in open-mmlab#970
• Fix repeated warning by @HAOCHENYE in open-mmlab#992
• Fix lint by @zhouzaida in open-mmlab#993
• Fix AMP in Ascend and support using NPUJITCompile environment by @luomaoling in open-mmlab#994
• Fix inferencer gets wrong configs path by @HAOCHENYE in open-mmlab#996

Docs

• Translate "Debug Tricks" to English by @enkilee in open-mmlab#953
• Translate "Model Analysis" document to English by @enkilee in open-mmlab#956
• Translate "Model Complexity Analysis" to Chinese. by @VoyagerXvoyagerx in open-mmlab#969
• Add a document about setting interval by @YuetianW in open-mmlab#964
• Translate "how to set random seed" by @xin-li-67 in open-mmlab#930
• Fix typo by @zhouzaida in open-mmlab#965
• Fix typo in hook document by @acdart in open-mmlab#980
• Fix changelog date by @HAOCHENYE in open-mmlab#986

Contributors

A total of 10 developers contributed to this release. Thanks @xin-li-67, @acdart, @enkilee, @YuetianW, @luomaoling, @nijkah, @VoyagerXvoyagerx, @zhouzaida, @HAOCHENYE, @C1rN09

v0.6.0 (02/24/2023)

Highlights

• Support Apex with ApexOptimWrapper
• Support analyzing model complexity.
• Add Lion optimizer.
• Support using environment variables in the config file.

New Features & Enhancements

• Support model complexity computation by @tonysy in open-mmlab#779
• Add Lion optimizer by @zhouzaida in open-mmlab#952
• Support Apex with ApexOptimWrapper by @xcnick in open-mmlab#742
• Support using environment variable in config file. by @jbwang1997 in open-mmlab#744
• Improve registry infer_scope by @zhouzaida in open-mmlab#334
• Support configuring timeout in dist configuration by @apacha in open-mmlab#877
• Beautify the print result of the registry by @Eiuyc in open-mmlab#922
• Refine the style of table by @zhouzaida in open-mmlab#941
• Refine the repr of Registry by @zhouzaida in open-mmlab#942
• Feature NPUProfilerHook by @luomaoling in open-mmlab#925
• Refactor hooks unittest by @HAOCHENYE in open-mmlab#946
• Temporarily fix collect_env raise errors and stops programs by @C1rN09 in open-mmlab#944
• Make sure Tensors to broadcast is contiguous by @XWHtorrentx in open-mmlab#948
• Clean the UT warning caused by pytest by @zhouzaida in open-mmlab#947

Bug fixes

• Backend_args should not be modified by get_file_backend by @zhouzaida in open-mmlab#897
• Support update np.ScalarType data in message_hub by @HAOCHENYE in open-mmlab#898
• Support rendering Chinese character in Visualizer by @KevinNuNu in open-mmlab#887
• Fix the bug of DefaultOptimWrapperConstructor when the shared parameters do not require the grad by @HIT-cwh in open-mmlab#903

Docs

• Add the document for the transition between IterBasedTraining and EpochBasedTraining by @HAOCHENYE in open-mmlab#926
• Introduce how to set random seed by @zhouzaida in open-mmlab#914
• Count FLOPs and parameters by @zhouzaida in open-mmlab#939
• Enhance README by @Xiangxu-0103 in open-mmlab#835
• Add a document about debug tricks by @zhouzaida in open-mmlab#938
• Refine the format of changelog and visualization document by @zhouzaida in open-mmlab#906
• Move examples to a new directory by @zhouzaida in open-mmlab#911
• Resolve warnings in sphinx build by @C1rN09 in open-mmlab#915
• Fix docstring by @zhouzaida in open-mmlab#913
• How to set the interval parameter by @zhouzaida in open-mmlab#917
• Temporarily skip errors in building pdf docs at readthedocs by @C1rN09 in open-mmlab#928
• Add the links of twitter, discord, medium, and youtube by @vansin in open-mmlab#924
• Fix typo shedule by @Dai-Wenxun in open-mmlab#936
• Fix failed URL by @zhouzaida in open-mmlab#943

Contributors

A total of 15 developers contributed to this release. Thanks @Eiuyc, @xcnick, @KevinNuNu, @XWHtorrentx, @tonysy, @zhouzaida, @Xiangxu-0103, @Dai-Wenxun, @jbwang1997, @apacha, @C1rN09, @HIT-cwh, @vansin, @HAOCHENYE, @luomaoling.

v0.5.0 (01/20/2023)

Highlights

• Add BaseInferencer to provide a general inference interface
• Provide ReduceOnPlateauParamScheduler to adjust learning rate by metric
• Deprecate support for Python3.6

New Features & Enhancements

• Deprecate support for Python3.6 by @HAOCHENYE in open-mmlab#863
• Support non-scalar type metric value by @mzr1996 in open-mmlab#827
• Remove unnecessary calls and lazily import to speed import performance by @zhouzaida in open-mmlab#837
• Support ReduceOnPlateauParamScheduler by @LEFTeyex in open-mmlab#819
• Disable warning of subprocess launched by dataloader by @HAOCHENYE in open-mmlab#870
• Add BaseInferencer to provide general interface by @HAOCHENYE in open-mmlab#874

Bug Fixes

• Fix support for Ascend device by @wangjiangben-hw in open-mmlab#847
• Fix Config cannot parse base config when there is . in tmp path, etc. tmp/a.b/c by @HAOCHENYE in open-mmlab#856
• Fix unloaded weights will not be initialized when using PretrainedIinit by @HAOCHENYE in open-mmlab#764
• Fix error package name defined in PKG2PROJECT by @HAOCHENYE in open-mmlab#872

Docs

• Fix typos in advanced_tutorials/logging.md by @RangeKing in open-mmlab#861
• Translate CN train_a_gan to EN by @yaqi0510 in open-mmlab#860
• Update fileio.md by @Xiangxu-0103 in open-mmlab#869
• Add Chinese documentation for inferencer. by @HAOCHENYE in open-mmlab#884

Contributors

A total of 8 developers contributed to this release. Thanks @LEFTeyex, @RangeKing, @yaqi0510, @Xiangxu-0103, @wangjiangben-hw, @mzr1996, @zhouzaida, @HAOCHENYE.

v0.4.0 (12/28/2022)

Highlights

• Registry supports importing modules automatically
• Upgrade the documentation and provide the English documentation
• Provide ProfileHook to profile the running process

New Features & Enhancements

• Add conf_path in PetrelBackend by @sunyc11 in open-mmlab#774
• Support multiple --cfg-options. by @mzr1996 in open-mmlab#759
• Support passing arguments to OptimWrapper.update_params by @twmht in open-mmlab#796
• Make get_torchvision_model compatible with torch 1.13 by @HAOCHENYE in open-mmlab#793
• Support flat_decay_mult and fix bias_decay_mult of depth-wise-conv in DefaultOptimWrapperConstructor by @RangiLyu in open-mmlab#771
• Registry supports importing modules automatically. by @RangiLyu in open-mmlab#643
• Add profiler hook functionality by @BayMaxBHL in open-mmlab#768
• Make TTAModel compatible with FSDP. by @HAOCHENYE in open-mmlab#611

Bug Fixes

• hub.get_model fails on some MMCls models by @C1rN09 in open-mmlab#784
• Fix BaseModel.to and BaseDataPreprocessor.to to make them consistent with torch.nn.Module by @C1rN09 in open-mmlab#783
• Fix creating a new logger at PretrainedInit by @xiexinch in open-mmlab#791
• Fix ZeroRedundancyOptimizer ambiguous error with param groups when PyTorch < 1.12.0 by @C1rN09 in open-mmlab#818
• Fix MessageHub set resumed key repeatedly by @HAOCHENYE in open-mmlab#839
• Add progress argument to load_from_http by @austinmw in open-mmlab#770
• Ensure metrics is not empty when saving best checkpoint by @zhouzaida in open-mmlab#849

Docs

• Add contributing.md by @HAOCHENYE in open-mmlab#754
• Add gif to 15 min tutorial by @HAOCHENYE in open-mmlab#748
• Refactor documentations and translate them to English by @zhouzaida in open-mmlab#786
• Fix document link by @MambaWong in open-mmlab#775
• Fix typos in EN contributing.md by @RangeKing in open-mmlab#792
• Translate data transform docs. by @mzr1996 in open-mmlab#737
• Replace markdown table with html table by @HAOCHENYE in open-mmlab#800
• Fix wrong example in Visualizer.draw_polygons by @lyviva in open-mmlab#798
• Fix docstring format and rescale the images by @zhouzaida in open-mmlab#802
• Fix failed link in registry by @zhouzaida in open-mmlab#811
• Fix typos by @shanmo in open-mmlab#814
• Fix wrong links and typos in docs by @shanmo in open-mmlab#815
• Translate save_gpu_memory.md by @xin-li-67 in open-mmlab#803
• Translate the documentation of hook design by @zhouzaida in open-mmlab#780
• Fix docstring format by @zhouzaida in open-mmlab#816
• Translate registry.md by @xin-li-67 in open-mmlab#817
• Update docstring of BaseDataElement by @Xiangxu-0103 in open-mmlab#836
• Fix typo by @Xiangxu-0103 in open-mmlab#841
• Update docstring of structures by @Xiangxu-0103 in open-mmlab#840
• Translate optim_wrapper.md by @xin-li-67 in open-mmlab#833
• Fix link error in initialize tutorial. by @HAOCHENYE in open-mmlab#843
• Fix table in initialized.md by @HAOCHENYE in open-mmlab#844

Contributors

A total of 16 developers contributed to this release. Thanks @BayMaxBHL, @RangeKing, @Xiangxu-0103, @xin-li-67, @twmht, @shanmo, @sunyc11, @lyviva, @austinmw, @xiexinch, @mzr1996, @RangiLyu, @MambaWong, @C1rN09, @zhouzaida, @HAOCHENYE

v0.3.2 (11/24/2022)

New Features & Enhancements

• Send git errors to subprocess.PIPE by @austinmw in open-mmlab#717
• Add a common TestRunnerTestCase to build a Runner instance. by @HAOCHENYE in open-mmlab#631
• Align the log by @HAOCHENYE in open-mmlab#436
• Log the called order of hooks during training process by @songyuc in open-mmlab#672
• Support setting eta_min_ratio in CosineAnnealingParamScheduler by @cir7 in open-mmlab#725
• Enhance compatibility of revert_sync_batchnorm by @HAOCHENYE in open-mmlab#695

Bug Fixes

• Fix distributed_training.py in examples by @PingHGao in open-mmlab#700
• Format the log of CheckpointLoader.load_checkpoint by @HAOCHENYE in open-mmlab#685
• Fix bug of CosineAnnealingParamScheduler by @fangyixiao18 in open-mmlab#735
• Fix add_graph is not called bug by @shenmishajing in open-mmlab#632
• Fix .pre-commit-config-zh-cn.yaml pyupgrade-repo github->gitee by @BayMaxBHL in open-mmlab#756

Docs

• Add English docs of BaseDataset by @GT9505 in open-mmlab#713
• Fix BaseDataset typo about lazy initialization by @MengzhangLI in open-mmlab#733
• Fix typo by @zhouzaida in open-mmlab#734
• Translate visualization docs by @xin-li-67 in open-mmlab#692

v0.3.1 (11/09/2022)

Highlights

• Fix error when saving best checkpoint in ddp-training

New Features & Enhancements

• Replace print with print_log for those functions called by runner by @HAOCHENYE in open-mmlab#686

Bug Fixes

• Fix error when saving best checkpoint in ddp-training by @HAOCHENYE in open-mmlab#682

Docs

• Refine Chinese tutorials by @Xiangxu-0103 in open-mmlab#694
• Add MMEval in README by @sanbuphy in open-mmlab#669
• Fix error URL in runner docstring by @HAOCHENYE in open-mmlab#668
• Fix error evaluator type name in evaluator.md by @sanbuphy in open-mmlab#675
• Fix typo in utils.md @sanbuphy in open-mmlab#702

v0.3.0 (11/02/2022)

New Features & Enhancements

• Support running on Ascend chip by @wangjiangben-hw in open-mmlab#572
• Support torch ZeroRedundancyOptimizer by @nijkah in open-mmlab#551
• Add non-blocking feature to BaseDataPreprocessor by @shenmishajing in open-mmlab#618
• Add documents for clip_grad, and support clip grad by value. by @HAOCHENYE in open-mmlab#513
• Add ROCm info when collecting env by @zhouzaida in open-mmlab#633
• Add a function to mark the deprecated function. by @HAOCHENYE in open-mmlab#609
• Call register_all_modules in Registry.get() by @HAOCHENYE in open-mmlab#541
• Deprecate _save_to_state_dict implemented in mmengine by @HAOCHENYE in open-mmlab#610
• Add ignore_keys in ConcatDataset by @BIGWangYuDong in open-mmlab#556

Docs

• Fix cannot show changelog.md in chinese documents. by @HAOCHENYE in open-mmlab#606
• Fix Chinese docs whitespaces by @C1rN09 in open-mmlab#521
• Translate installation and 15_min by @xin-li-67 in open-mmlab#629
• Refine chinese doc by @Tau-J in open-mmlab#516
• Add MMYOLO link in README by @Xiangxu-0103 in open-mmlab#634
• Add MMEngine logo in docs by @zhouzaida in open-mmlab#641
• Fix docstring of BaseDataset by @HAOCHENYE in open-mmlab#656
• Fix docstring and documentation used for hub.get_model by @zengyh1900 in open-mmlab#659
• Fix typo in docs/zh_cn/advanced_tutorials/visualization.md by @MambaWong in open-mmlab#616
• Fix typo docstring of DefaultOptimWrapperConstructor by @triple-Mu in open-mmlab#644
• Fix typo in advanced tutorial by @cxiang26 in open-mmlab#650
• Fix typo in Config docstring by @sanbuphy in open-mmlab#654
• Fix typo in docs/zh_cn/tutorials/config.md by @Xiangxu-0103 in open-mmlab#596
• Fix typo in docs/zh_cn/tutorials/model.md by @C1rN09 in open-mmlab#598

Bug Fixes

• Fix error calculation of eta_min in CosineRestartParamScheduler by @Z-Fran in open-mmlab#639
• Fix BaseDataPreprocessor.cast_data could not handle string data by @HAOCHENYE in open-mmlab#602
• Make autocast compatible with mps by @HAOCHENYE in open-mmlab#587
• Fix error format of log message by @HAOCHENYE in open-mmlab#508
• Fix error implementation of is_model_wrapper by @HAOCHENYE in open-mmlab#640
• Fix VisBackend.add_config is not called by @shenmishajing in open-mmlab#613
• Change strict_load of EMAHook to False by default by @HAOCHENYE in open-mmlab#642
• Fix open encoding problem of Config in Windows by @sanbuphy in open-mmlab#648
• Fix the total number of iterations in log is a float number. by @jbwang1997 in open-mmlab#604
• Fix pip upgrade CI by @HAOCHENYE in open-mmlab#622

New Contributors

• @shenmishajing made their first contribution in open-mmlab#618
• @Xiangxu-0103 made their first contribution in open-mmlab#596
• @Tau-J made their first contribution in open-mmlab#516
• @wangjiangben-hw made their first contribution in open-mmlab#572
• @triple-Mu made their first contribution in open-mmlab#644
• @sanbuphy made their first contribution in open-mmlab#648
• @Z-Fran made their first contribution in open-mmlab#639
• @BIGWangYuDong made their first contribution in open-mmlab#556
• @zengyh1900 made their first contribution in open-mmlab#659

v0.2.0 (10/11/2022)

New Features & Enhancements

• Add SMDDP backend and support running on AWS by @austinmw in open-mmlab#579
• Refactor FileIO but without breaking bc by @zhouzaida in open-mmlab#533
• Add test time augmentation base model by @HAOCHENYE in open-mmlab#538
• Use torch.lerp\_() to speed up EMA by @RangiLyu in open-mmlab#519
• Support converting BN to SyncBN by config by @HAOCHENYE in open-mmlab#506
• Support defining metric name in wandb backend by @okotaku in open-mmlab#509
• Add dockerfile by @zhouzaida in open-mmlab#347

Docs

• Fix API files of English documentation by @zhouzaida in open-mmlab#525
• Fix typo in instance_data.py by @Dai-Wenxun in open-mmlab#530
• Fix the docstring of the model sub-package by @zhouzaida in open-mmlab#573
• Fix a spelling error in docs/zh_cn by @cxiang26 in open-mmlab#548
• Fix typo in docstring by @MengzhangLI in open-mmlab#527
• Update config.md by @Zhengfei-0311 in open-mmlab#562

Bug Fixes

• Fix LogProcessor does not smooth loss if the name of loss doesn't start with loss by @liuyanyi in open-mmlab#539
• Fix failed to enable detect_anomalous_params in MMSeparateDistributedDataParallel by @HAOCHENYE in open-mmlab#588
• Fix CheckpointHook behavior unexpected if given filename_tmpl argument by @C1rN09 in open-mmlab#518
• Fix error argument sequence in FSDP by @HAOCHENYE in open-mmlab#520
• Fix uploading image in wandb backend @okotaku in open-mmlab#510
• Fix loading state dictionary in EMAHook by @okotaku in open-mmlab#507
• Fix circle import in EMAHook by @HAOCHENYE in open-mmlab#523
• Fix unit test could fail caused by MultiProcessTestCase by @HAOCHENYE in open-mmlab#535
• Remove unnecessary "if statement" in Registry by @MambaWong in open-mmlab#536
• Fix _save_to_state_dict by @HAOCHENYE in open-mmlab#542
• Support comparing NumPy array dataset meta in Runner.resume by @HAOCHENYE in open-mmlab#511
• Use get instead of pop to dump runner_type in build_runner_from_cfg by @nijkah in open-mmlab#549
• Upgrade pre-commit hooks by @zhouzaida in open-mmlab#576
• Delete the error comment in registry.md by @vansin in open-mmlab#514
• Fix Some out-of-date unit tests by @C1rN09 in open-mmlab#586
• Fix typo in MMFullyShardedDataParallel by @yhna940 in open-mmlab#569
• Update Github Action CI and CircleCI by @zhouzaida in open-mmlab#512
• Fix unit test in windows by @HAOCHENYE in open-mmlab#515
• Fix merge ci & multiprocessing unit test by @HAOCHENYE in open-mmlab#529

New Contributors

• @okotaku made their first contribution in open-mmlab#510
• @MengzhangLI made their first contribution in open-mmlab#527
• @MambaWong made their first contribution in open-mmlab#536
• @cxiang26 made their first contribution in open-mmlab#548
• @nijkah made their first contribution in open-mmlab#549
• @Zhengfei-0311 made their first contribution in open-mmlab#562
• @austinmw made their first contribution in open-mmlab#579
• @yhna940 made their first contribution in open-mmlab#569
• @liuyanyi made their first contribution in open-mmlab#539