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 2.2.02Jun129May282524228732130Apr2922212019181716854229Mar21191817161413118765128Feb272624222019161413129875229Jan242322191722Dec141319Sep518Aug17161514865432131Jul3028272622Jun22May1615121110[tests/data/histograms] Pretty formatRemove .gitmodulesMake code and tests flexible to the use of a pre-processed Montgomery dataset to speed-up testing where possible (closes #74)Remove unused test data (closes #75)Merge branch 'readthedocs' into 'main'[doc] Host documentation on readthedocs.orgMerge branch 'upload-moretraininfo' into 'main'Merge branch 'update-lockfile' into 'main'Merge branch 'lightning-acc' into 'main'[scripts.train] Fix description[pyproject] Fixes unresolvable cuda environment making nvidia channel more prioritaryFix qa. #69Update reference to batch accumulation flag. #69Change flag for batch accumulation. #25Fix qa #69.Add more train info into experiment table and relocate trainlog.pdf path. #69Fix typo on calling sub-field. #25Fixed help text for new usage. #25Splitting into chunks is removed as lightning supports it anymore. #25Merge branch 'issue-70' into 'main'[tests.test_cli] Fix default output directory for train analysis[engine.callbacks] Be more specific about the type of missing logs[scripts.train_analysis] Add indication of lowest validation loss epochs on loss plots (closes #70)[tests/data/histograms] Pretty format[pre-commit] Update and add further rules; clean-up[pixi] Update lockfile[pyproject] Remove dev deps from test-ci-alternative[pyproject] Use python-version-agnostic environment name[ci] Also install libopengl0 via apt[dockerfile] Add installation of libegl1[ci] Also install libegl1 via aptMerge branch 'loss-balancing' into 'main'[pyproject] Depend on matplotlib-base for conda packages[trainer] Hardcode name of loss to monitor[logging] Prevent PL from adding dataloader index to logged metrics[model] Use a single type of loss for train and validation[model] Fix detection of binary targets in loss balancing[doc] Add base Model to documentation[model] Model takes loss type and arguments during instanciation[model] Fix balancing of multiclass targets