Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
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
Created with Raphaël 2.2.010Jul9432128Jun26242113107432129May282524228732130Apr2922212019181716854229Mar21191817161413118765128Feb272624222019161413129875229Jan242322191722Dec141319Sep518Aug171615[models.transforms] Make RGB transform work for TVTensors[doc] Fix doc building[classify.scripts.saliency] Incorporate evaluation code directly on interpretability/completeness analysis; Remove outdated saliency-evaluation script and engine; Fix test units[classify.scripts.saliency.completeness] Add basic analysis and plots[classify.scripts.saliency.interpretability] Fix comments and improve variable descriptors[classify.engine.saliency.interpretability] Implement filter for operating on specific datasets only; Implement basic analysis for interpretability[classify.engine.saliency] Implement filter for operating on specific datasets only[tests] Update test histograms[tests] Remove outdated tests[saliency] Remove unused iou code[classify.scripts.saliency] Simplify script outputs[classify.scripts.saliency] Improve documention[models.transforms] Make RGB transform work correctly[ignore] Ignore more CI files[helpers] Fix test histograms generation[ignore] Ignore CI files[pre-commit] Revert version bump on reuse to avoid deprecation warning for nowFix reuse warningsMerge branch 'simplify' into 'common-package'Simplify package organisation[libs.segmentation.models.losses] Use explicit version of super() to solve issues with Python-3.11[helpers] Update scripts to generate reference histograms[tests] Update reference histograms[tests.test_transforms] Remove unused test[libs.common.models.transforms] Maximise use of torchvision transforms; Closes #86 as well concluding previous commit[config.models] Restore model transforms; Unify transform strategies[libs.segmentation.engine.evaluator] Fix precision-recall estimates when precision and recall == 0[libs.segmentation.models.losses] Simplify loss calculation; Specialize prediction step on lwnet[libs.common.engine.callbacks] Fix access to last learning-rate[libs.segmentation.config.models.lwnet] Set default cosine annealing strategy to be closer to original publication[libs.common.engine.callbacks] Remove call to pdb[libs.common.engine.callbacks] Use appropriate interface for extracting the current learning rate from schedulers[evaluation] Homogenize segmentation/classification code; Use more credible constructs (DRY); Fix test units[libs.common.models.model] Do not assume pos_weight is a scalar[*/models] Unify modelling by applying DRY; Fix a nasty loss-balancing bug; Make loss-configuration private to base model class; Automate loss-configuration; Fix typing across model submodules; Move `name` into parent type[various] Remove retrieval of "mednet" logger in favour of the modules[various] Fix multiple import errors after rebase[pyproject] Reset correct module path for visceral[doc] Fix doc building[libs.classification.config.models.cnn3d] Fix qa