Skip to content
Snippets Groups Projects
Select Git revision
  • aifmd_dataset_support
  • main default protected
  • robertos-branch
  • v1.3.1
  • v1.3.0
  • v1.2.0
  • v1.1.0
  • v1.0.2
  • v1.0.1
  • v1.0.0
  • v1.0.0b0
  • v1.0.0a0
12 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.05Sep432130Aug232221207611Jul109432128Jun26242113107432129May282524228732130Apr2922212019181716854229Mar21191817161413118765128Feb272624222019161413129875229Jan24[models.losses] Improve calculation of loss weightsMerge branch 'issue-68-multi-label' into 'main'Implements multi-label support for BCEWithLogitsLoss and MOONBCEWithLogitsLoss[data.segment.drionsdb] Fix implementation after refactoring[models.losses] Centralize all custom losses in a single place[scripts.classify.evaluate] By default, do not calculate credible regions (slow on large datasets)[data.classify.nih_cxr14] Fix file access[data.datamodule] Always set the mp context to "spawn"[data.datamodule] Update issue link on warning[data.datamodule] Use torch.multiprocessing instead of plain one[tests] Add test for cached dataset[data.datamodule] Fix comment[data.datamodule] Disables multiprocessing dataset caching on Linux[data.datamodule] Use spawn on linux as well[tests.test_transforms] Streamline tests[tests.test_transforms] Minor adjustment at reference after to_tensor() deprecation[data.classify,data.segment] Avoid to_tensor() torchvision deprecation[pyproject] Downgrade pytorch to avoid https://github.com/conda-forge/pytorch-cpu-feedstock/issues/254[data.classify.nih_cxr14] Avoid auto-conversion to float64 tensors[models.classify.densenet,alexnet] Fix imagenet normalizer [ci skip]Increased latest version to 1.1.1b0 [skip ci]Increased stable version to 1.1.0v1.1.0v1.1.0[doc] Improve data-model descriptionMerge branch 'support-multi-class' into 'main'Support for multi-label classification[rtd] Also fetch tags when building on RTD.org [ci skip]Increased latest version to 1.0.3b0 [skip ci]Increased stable version to 1.0.2v1.0.2v1.0.2[scripts.info] Print supported databases even if no RC foundIncreased latest version to 1.0.2b0 [skip ci]Increased stable version to 1.0.1v1.0.1v1.0.1[pyproject] Relax constraints to the maximum[scripts.classify.saliency.view] Fix main script documentation[README] Make basic package description more uniform across different instances; Independent of a particular taskMerge branch 'issue-88-pytorch-upgrade' into 'main'[models.transforms] Use RGB transform from torchvision 0.18.1 (closes #91)[tests/classify] Remove unused file[tests.classify.test_montgomery] Simplify model tests[tests/classify] Rectify tests depending on the RGB() transform output after upgrade to torchvision 0.18.1[pyproject] Re-add pip as a cuda env dependency for docker installation