Skip to content
Snippets Groups Projects
Select Git revision
  • main default protected
  • v0.10.4
  • v0.10.3
  • v0.10.2
  • v0.10.1
  • v0.10.0
  • v0.9.1
  • v0.9.0
  • v0.8.5
  • v0.8.4
  • v0.8.3
  • v0.8.2
  • v0.8.1
  • v0.8.0
  • v0.7.4
  • v0.7.3
  • v0.7.2
  • v0.7.1
  • v0.7.0
  • v0.6.0
  • v0.5.0
21 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.012Apr11107654330Mar2928261917161514131210987652127Feb2423222120191615141387653131Jan29201613121096430Dec282726232221201916131211108765429Nov25242322211918171598632131Oct2827262421191817161514131211109826Sep221915141397131Aug302928262524232221191817161513119842130Jul272221201915141312865430Jun2928272423222117161514131098765131May302726252421201918171312106529Apr27262524222120151285131Mar30292826251413121098765432127Feb26252422[Docs] Translate the migration/hook.md to English (#1054)[Fix] Fix publish multiple checkpoints when using multiple GPUs (#1059) (#1070)[Docs] Introduce the use of wandb and tensorboard (#912)[Enhance] Complement type hint of get_model_complexity_info() (#1064)[Feature] Support registering partial functions and more (#595)[Enhance] Make the parameters of get_model_complexity_info() friendly (#1056)[Enhance] Allow users to customize worker_init_fn of Dataloader (#1038)[Enhance] Remove useless variable declaration (#1052)[Enhancement] Support BoolTensor and LongTensor on Ascend NPU (#1011)[Feature] Add vis backend for MLflow. (#878)[Refactor] Refactor logger hook unit tests (#797)Bump version to v0.7.2 (#1051)v0.7.2v0.7.2[Refactor] Refactor checkpointhook unit tests (#789)[Fix] Initialize nested modules in ddp which define 'init_weights' method (#1045)[Fix] Update the ClassBalancedDataset logic to keep len(repeat_factors) = len(dataset) (#1048)[Enhancement] Align the evaluation result in log (#1034)[Enhance] Suppress registry warning (#1043)v0.7.1v0.7.1Bump version to 0.7.1 (#1007)[Enhance] Make sure the FileHandler still alive after `torch.compile` (#1021)[Enhancement] Do not print config if it is empty (#1028)[Enhance] Do no catch unnecessary exceptions in registry (#1010)Skip test_dist in Github Action (#1027)[Fix] Add PyTorch 2.0 CI and fix unit tests (#1026)[Feature] Publish models after training if published_keys is set in CheckpointHook (#987)[Docs] Add more social networking links (#1024)[Fix] Fix there is no space between data_time and metric (#1025)[Docs] Minor fixes in docs to remove or replace unicode chars with ascii chars (#1018)[Refactor] Refactor the `get_device` and `is_npu_available` (#1004)[Enhancement] Unify the use of print_log and logger (#997)[Enhance] Add `mmpretrain` to the `MODULE2PACKAGE` (#1002)Bump version to v0.7.0 (#999)v0.7.0v0.7.0[Fix] Fix pytorch2.0 does not support sys._getframe used to remove private keys (#979)[Fix] Fix inferencer gets wrong configs path (#996)Add citation in README (#995)[Fix] Fix AMP in Ascend and support using NPUJITCompile environment (#994)fix lint (#993)[Enhancement] Support dumping logs of different ranks (#968)[Docs] Translate "Model Complexity Analysis" to Chinese (#969)[Enhancement] Support writing data to `vis_backend` with prefix (#972)[Feature] Support torch.compile since PyTorch2.0 (#976)

Datenschutz und Impressum