galaxy-ml - Release v0.10.0

Changes

• Updates scikit-learn to 1.1.x, tensorflow to v2.10.x, xgboost to 1.7.x.
• Add github workflow to test tools using docker container as dependencies.
• Restructure search algos in searchCV tool.
• Serialize html repr for models.
• Test utils and tools in python 3.9.

Bug Fixes

• Fix some tool lint errors
• Fix lots of flake8 errors but ignoring F401,F821, I100 and W503.

- HTML
Published by qiagu over 3 years ago

galaxy-ml - V0.9.0

Changes

• Updates scikit-learn to v0.24.x, tensorflow to v2.4.x, xgboost to v1.3.x, mlxtend to v0.17.x, skrebate to v.62, imbalanced-learn to v0.8.x and so on.
• Makes loadmodel and savemodel util methods in kerasgalaxymodel module.
• Refactors _SafePickler and moves it from utils to model_persist.
• Refactors dump_model_to_h5 and load_model_from_h5 to dynamically save and load xgboost and tensorflow models.
• Replaces pickled models with h5mlm models in all tools.

Bug Fixes

- HTML
Published by qiagu almost 5 years ago

galaxy-ml - V0.8.3

Changes

• Makes image deep learning tool.
• Adds BayesSearchCV in searchcv tool.
• Makes new metric spearman_correlation_score and its corresponding scorer.
• Adds a new parameter n_stratification_bins to OrderedKFold and RepeatedOrderedKFold.
• Adds multi-output classification support for KerasGClassification and KerasGBactchClassifier.

Bug Fixes

• Fixes dependencies issue, numpy < 1.19, h5py < 3, tensorflow-estimator==1.15.1.
• Fixes various test issues.

- HTML
Published by qiagu about 5 years ago

galaxy-ml - v0.8.1

Version 0.8.1 / tool_main: 1.0.8.1 / keras: 0.5.0 (12-12-2019)

Changes

• Changes all predict_score to decision_function.

Bug Fixes

• Fixes _SafePickler doesn't recognize binarize_target module.
• Fixes GridSearchCV has no predict_score issue.

- HTML
Published by qiagu over 6 years ago

galaxy-ml - V0.8.0

Version 0.8.0 / tool_main: 1.0.8.0 / keras: 0.5.0 (12-10-2019)

New Features

• Adds circleci config for both api and tool tests.
• Adds traintestsplit tool which supports shufflesplit, stratifiedshufflesplit, groupshufflesplit and orderedtarget split.
• Adds fittedmodeleval tool.
• Refactors binarize target estimators. There are a lot of improvements. One of them is that the estimator family now support most sklearn scorers.
• Adds clean_params in utils
• Adds cvresults outputs for nested inner CV and unfitted searchCV object from searchCV tool.
• Adds keras training and evaluation tool.
• Adds support of decision_function for binarize target classifiers.
• Adds matplotlib svg format option in ml_visualization_ex tool.
• Adds 'sklearn.ensemble.HistGradientBoostingClassifier' and 'sklearn.ensemble.HistGradientBoostingRegressor'
• Adds new regression scorer max_error.
• Upgade scikit-lean to v0.21.3, mlxtend to v0.17.0, imbalanced-learn to v0.5.0, keras to v2.3.1 and tensorflow to v1.15.0.

Changes

• Replaces all generators' fit with set_processing_attrs.
• Raises ValueError instead of [0, 1] normalization when predictions from BinarizeTargetRegressor go out of range.
• Refactors iraps_classifier module. Binarize target estimators do the same prediction as the wrapped estimator. A delicated predict_score is made to work with binarize scorers.
• Changes precision-recall curve and ROC curve to take headers and upgrade plotly to v4.3.0 in mlvisualizationex tool
• Changes to dynamic output of pipeline or final main estimator

Bug Fixes

• Fixes randomstate error in `predict_generator`.
• Fixes stale path issue by replacing relative paths with full paths.

- HTML
Published by qiagu over 6 years ago

galaxy-ml - V0.7.13

Version 0.7.13 / tool_main: 1.0.7.12 / stacking: 0.2.0 / keras: 0.4.2 (09-17-2019)

New Features

• Adds searchcv tools to output weights for deep learning models.
• Makes KerasGBatchClassifier.evalue to support multi-class and multi-label classification problem.
• Adds parameter verbose in KerasG models to output device placement.
• Adds metrics in keras model building tools.
• Makes train_test_eval tool.
• Makes GenomicVariantBatchGenerator.
• Makes model_prediction tool to support vcf file type.
• Adds plotly plotting tool facility for feature_importances, learning_curve, pr_curve and roc_curve.
• Adds _predict_generator to output y_true together with prediction results.
• Adds support of return_train_score for KerasGBatchClassifier in gridsearchcv.
• Adds ml_visualization.xml tool support many plots.

Changes

• Changes dependency tensorflow to tensorflow-gpu.
• Moves all tools to folder tools.
• Makes sklearn.preprocessing.Imputer deprecated.
• Updates dependencies in requrements.txt.
• Refactor keras_model_config tool by grouping layer key words arguments.
• Refactor the preprocessors.py into folder structure.

Bug Fixes

• Fixes KerasGBatchClassifier doesn't work with callbacks.
• Fixes GenomicIntervalBatchGenerator doesn't work in nested model validation.
• Fixes GenomicIntervalBatchGenerator failed for sequences in blacklist matches.

- HTML
Published by qiagu over 6 years ago