v0.4.11

What's Changed

• Add CaiT and XCiT models by @Animatory in #1
• Fixes & Improvements from EORA team by @VladVin in #6
• Re-engineered basic folders structure by @VladVin in #10
• create feature-classification_dataset by @VladislavPatrushev in #12
• feature_unsupervised_contrastive_dataset by @VladislavPatrushev in #13
• registry write by @PososikTeam in #14
• Feature retrieval dataset by @VladislavPatrushev in #15
• add flake8 by @VladislavPatrushev in #17
• change api base dataset by @VladislavPatrushev in #27
• Feature metric manager by @PososikTeam in #11
• Constructor by @VladVin in #22
• create feature hooks base model by @PososikTeam in #19
• Fixed order names of tests for JointLoss by @VladVin in #29
• feature-representation_metric by @PososikTeam in #16
• add base task by @VladislavPatrushev in #18
• Feature ResNet, SEResNet by @VladislavPatrushev in #26
• Feature dataset cifar10 by @VladislavPatrushev in #32
• Feature ResNet add urls by @VladislavPatrushev in #34
• Config structure by @PososikTeam in #25
• Hot FIX Metric test by @PososikTeam in #30
• Feature arc face head by @VladislavPatrushev in #21
• Feature base task bug-fix by @VladislavPatrushev in #33
• SOP run by @PososikTeam in #36
• Added new requierements by @VladVin in #37
• CIFAR-10 run by @VladVin in #38
• Updated README for the new concept by @VladVin in #40
• Feature HRNet by @VladislavPatrushev in #41
• Feature Segmentation Dataset by @VladislavPatrushev in #35
• Sphinx documentation by @VladVin in #43
• Set joint loss and optimizer as optional by @PososikTeam in #50
• featue save_onnx_ckpt by @VladislavPatrushev in #49
• Updated all dependencies and Dockerfile by @VladVin in #55
• Fixed logging steps in test configs by @VladVin in #58
• SOP training by @PososikTeam in #53
• Load checkpoint by @PososikTeam in #39
• hotfix-classification_task by @VladislavPatrushev in #66
• BaseModel class by @PososikTeam in #57
• MLFlow Logger by @PososikTeam in #54
• MLFlow config by @PososikTeam in #67
• Config MLFlow by @PososikTeam in #68
• hotfix-onnx_save_checkpoint_callback by @VladislavPatrushev in #70
• TorchOk packaging by @VladVin in #73
• SwinV2 by @PososikTeam in #47
• Segmentation by @PososikTeam in #71
• Adapt models from TIMM by @Animatory in #76
• Add efficientnet architecture by @Animatory in #87
• Add mobilenetv3 backbone by @Animatory in #88
• feature-onnxtask by @VladislavPatrushev in #48
• Add notebooks by @PososikTeam in #89
• Entrypoints by @PososikTeam in #77
• hotfix_multilog_dir by @VladislavPatrushev in #90
• Feature DaViT by @VladislavPatrushev in #56
• Freeze by @PososikTeam in #75
• feature-Unet by @VladislavPatrushev in #52
• Merge Hydra and PytorchLightning logging by @Animatory in #104
• Feature-forward_onnx by @VladislavPatrushev in #100
• Fixed deps for Python 3.7 by @VladVin in #105
• Global refactor by @Animatory in #107
• Fix Hydra logging timestamp by @Animatory in #108
• Add Beit backbone by @Animatory in #109
• Hotfix warnings by @VladislavPatrushev in #113
• Hotfix flake8 by @VladislavPatrushev in #111
• Feature github actions by @VladislavPatrushev in #112
• Refactor datasets by @Animatory in #115
• Hot fix for run all configs without errors by @PososikTeam in #116
• Feature_metric_with_utils by @VladislavPatrushev in #117
• Feature detection dataset by @PososikTeam in #118
• Hotfix python3.7 maintain by @VladislavPatrushev in #119
• Feature Pairwise Task by @VladislavPatrushev in #110
• Fix flake8 action by @VladislavPatrushev in #120
• fixed metric tests by @PososikTeam in #125
• Update for new lightning version by @Animatory in #127
• Fixed pandas version by @VladVin in #128
• Fixed numpy version by @VladVin in #129
• support many validation dataloaders by @PososikTeam in #130
• Fix COCO by @PososikTeam in #133
• implement bce loss by @PososikTeam in #131
• Dict support in MetricManager by @PososikTeam in #123
• Fix bugs by @PososikTeam in #126
• Fix dependencies in pyproject.toml by @VladVin in #134
• tests fix by @PososikTeam in #136
• Fix library versions by @PososikTeam in #137
• Full CI/CD: lint, test, publish by @VladVin in #135
• Fix additional tests by @PososikTeam in #138
• CI/CD tests fix and package version update by @VladVin in #139
• Fixed PyPI publishing trigger in CI/CD by @VladVin in #140
• Split CI/CD workflows by @VladVin in #141
• Update prerelease version by @VladVin in #142
• Remove uploading to TestPyPI from CD process & package update by @VladVin in #143

New Contributors

• @Animatory made their first contribution in #1
• @VladislavPatrushev made their first contribution in #12
• @PososikTeam made their first contribution in #14

Full Changelog: https://github.com/eora-ai/torchok/commits/v0.4.11