quadra

What's Changed

• feat: Align classification results with sklearn classification ones by @AlessandroPolidori in https://github.com/orobix/quadra/pull/63

Full Changelog: https://github.com/orobix/quadra/compare/v1.2.3...v1.2.4

What's Changed

• Hotfix/patch-datamodule-error by @lorenzomammana in https://github.com/orobix/quadra/pull/61

Full Changelog: https://github.com/orobix/quadra/compare/v1.2.2...v1.2.3

What's Changed

• Feature/upgrade anomalib by @lorenzomammana in https://github.com/orobix/quadra/pull/60

Full Changelog: https://github.com/orobix/quadra/compare/v1.2.1...v1.2.2

What's Changed

• feat: Automatically compute number of classes for classification tasks by @AlessandroPolidori in https://github.com/orobix/quadra/pull/56

Full Changelog: https://github.com/orobix/quadra/compare/v1.2.0...v1.2.1

What's Changed

• Feature/save raw outputs by @ebernatene in https://github.com/orobix/quadra/pull/37
• Fix/model wrapper torchscript by @lorenzomammana in https://github.com/orobix/quadra/pull/43
• Feature/export onnx models by @lorenzomammana in https://github.com/orobix/quadra/pull/40
• Feature/aig 1504 logical anomaly by @AlessandroPolidori in https://github.com/orobix/quadra/pull/45
• Bugfix/git hyperparameter crash by @lorenzomammana in https://github.com/orobix/quadra/pull/46
• Hotfix/test failures dev by @lorenzomammana in https://github.com/orobix/quadra/pull/49
• Feature/update mkdocs by @rcmalli in https://github.com/orobix/quadra/pull/48
• Feature/acknowledgement by @rcmalli in https://github.com/orobix/quadra/pull/51
• Feature/aig 1507 by @AlessandroPolidori in https://github.com/orobix/quadra/pull/50
• Feature/improve hash docs by @rcmalli in https://github.com/orobix/quadra/pull/52
• Release v1.2.0 by @rcmalli in https://github.com/orobix/quadra/pull/53
• fix: bump version to 1.2.0 by @rcmalli in https://github.com/orobix/quadra/pull/54

New Contributors

• @ebernatene made their first contribution in https://github.com/orobix/quadra/pull/37

Full Changelog: https://github.com/orobix/quadra/compare/v1.1.3...v1.2.0

What's Changed

• Fix/wrong patch extraction by @lorenzomammana in https://github.com/orobix/quadra/pull/42

Full Changelog: https://github.com/orobix/quadra/compare/v1.1.2...v1.1.3

What's Changed

• fix: Use best model checkpoint if available by default on lightning task test by @lorenzomammana in https://github.com/orobix/quadra/pull/39

Full Changelog: https://github.com/orobix/quadra/compare/v1.1.1...v1.1.2

What's Changed

• Fix tutorial link in documentation by @rcmalli in https://github.com/orobix/quadra/pull/36

Full Changelog: https://github.com/orobix/quadra/compare/v1.1.0...v1.1.1

What's Changed

• add: Migrate vit gradcam code + pytorch export format for classificat… by @AlessandroPolidori in https://github.com/orobix/quadra/pull/7
• Fix/binary linear model wrapper fix by @AlessandroPolidori in https://github.com/orobix/quadra/pull/12
• Bug: non squared input images break SklearnClassification task by @AlessandroPolidori in https://github.com/orobix/quadra/pull/15
• Docs/add newformat classif tutorial by @AlessandroPolidori in https://github.com/orobix/quadra/pull/18
• Fix/print config in anomaly base by @AlessandroPolidori in https://github.com/orobix/quadra/pull/19
• add: Add patch classification to tasks.md by @AlessandroPolidori in https://github.com/orobix/quadra/pull/24
• Feature/landing page by @rcmalli in https://github.com/orobix/quadra/pull/23
• Refactor/evaluation task by @AlessandroPolidori in https://github.com/orobix/quadra/pull/22
• Feature/experiment manager by @lorenzomammana in https://github.com/orobix/quadra/pull/3
• Feature/documentation Github Actions by @rcmalli in https://github.com/orobix/quadra/pull/29
• Fix/network builder kwargs by @lorenzomammana in https://github.com/orobix/quadra/pull/30
• fix: update readme links to always latest by @rcmalli in https://github.com/orobix/quadra/pull/32
• fix: update image links with correct relative path by @rcmalli in https://github.com/orobix/quadra/pull/33

Full Changelog:

Added

• Add ModelManager class to manage model deployments on model tracking platforms (e.g. MLFlow).
• Add automatic storage of file hashes in BaseDataModule class for better experiment reproducibility and tracking.
• Automatically load transforms parameters from model info file in base Evaluation task.
• Add support for vit explainability using Attention Gradient Rollout method.
• Add export of pytorch model for Classification and SklearnClassification tasks.
• Add automatical detection of model input shapes for better exportation capabilities. Add support for custom input shapes like models with multiple inputs.
• Add documentation landing page, improve colore themes and logo.
• Add github actions to automatically build and deploy documentation of main and dev PRs.

Changed

• Refactor evaluation task to be more generic and improve inheritance capabilities.
• Refactor export_types parameter to export for better configurability of export parameters.
• Change input_size model info parameter from HXW to a list of actual model parameters for inference (e.g [(3, 224, 224)]).

Fixed

• Fix gradcam not working properly with non rectangular images.
• Fix logistic regression wrapper not working properly with 2 classes for torch classification.
• Fix wrong typings in NetworkBuilder's init method.
• Fix broken links in documentation.
• Fix minor documentations issues.

What's Changed

Fix circular import making some tests not running properly by @lorenzomammana in https://github.com/orobix/quadra/pull/28

What's Changed

Fixed anomaly detection training not working with non convential image extensions (e.g. .BMP)

What's Changed

• Feature/default experiment config by @lorenzomammana in https://github.com/orobix/quadra/pull/1
• Feature/packaging by @rcmalli in https://github.com/orobix/quadra/pull/4
• Fix/readme badges by @rcmalli in https://github.com/orobix/quadra/pull/6
• fix: SCL inherits from Task, not from LightningTask by @AlessandroPolidori in https://github.com/orobix/quadra/pull/8

This release aims to target an error on the segmentation datamodules for which the validation dataset was used in place of the train dataset.