Skip to content
Snippets Groups Projects
Select Git revision
  • 1.1.x
  • main default protected
  • v1.5.0
  • v1.4.0
  • v1.3.0
  • v1.2.0
  • v1.1.1
  • models-v1
  • v1.1.0
  • v1.0.1
  • v1.0.0
11 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.029Jun2624527May20191815131211865130Apr2928272625242322212017151413121110987543223Mar2119171618Feb14Nov30Oct1617Sep1210965230Aug2928261914131231Jul251815985228Jun27252419629May242322161514109129Apr252418[engine.evaluator] Fix function to calculate patch performance[engine.evaluator] Dump scores for patches as well[script] Add significance test and support for it on evaluation[doc] Update baseline results[models.driu_od] Fix copy-n-paste error on backbone layer selection[test.test_cli] Fix unit test[engine,script] Simplify implementation of subset folder structure; Fix unit tests[script.*] Change prediction to write dataset predictions on a subfolder named after the dataset name in the current run[configs.datasets.hrf.xtest] Fix dictionary key[configs.datasets.hrf] Add full-resolution train/test sets as evaluation datasets[conda] Relax pytorch on OSX requirements1.1.x1.1.x[engine.inferencer] Fix type mismatch in pos/neg operations[modeling.losses] Patch weak_script hack[configs.datasets] Fix stare/chasedb1 covd configurations to avoid warning[script.analyze] Issue warning if keys do not match on for second-annotator evaluations (should not happen, normally)[script.analyze] Insert safey checks on automated analysis script[script.evaluate] Fix second-annotator comparisons with a key-resemblance check; Add assertions in engine.evaluator to ensure proper comparisons[readme] Set coverage to point to coverage report in docs [ci skip][test.test_models] Change mode to 0644[models.normalizer] Change mode to 0644Merge branch 'model-cleanup' into 'master'[test.test_summary] Fix resunet test[test.test_summary] Fix imports[utils.checkpointer] Remove custom serialization[models,configs/models] Cleanup model implementation to re-use all backbones from torchvision; Simplify checkpointer; Remove own implementation of model_zoo; Implement normalization for torchvision-based backbones; Enables pytorch installation on osx[test.test_config] Check image boundsmodels-v1models-v1[doc/scripts/baselines] Fix script[doc/scripts] Reformat baselines.sh [ci skip][doc] Reset scripts for baseline and xtest [ci skip][test.test_config] Fix config tests after db remodelling[configs.datasets] Use "validation" as key for default validation set[script.experiment] Use, preferably, the model with lowest validation error[engine.predictor] Use non-blocking operation to predictor speed-up[engine.ssltrainer] Validation does not compute SSL loss, just standard model performance[engine.*trainer] Set non-blocking operation for CPU->GPU data transfers to make communication asynchronous[engine] More comments on model.train() and model.to() usage[engine.*trainer] Optimize validation during training with torch.no_grad() and model.eval()[configs.datasets.*.covd] Add test data from COVD-target datasets as training data[engine.ssltrainer] Make logged variable names more consistent[configs.datasets.*.covd] Fix validation set