Model Tuner 0.0.24a

Model Tuner Version 0.0.24a Changelog

• Updated .gitignore to incl. doctrees

• Added pickleObjects tests and updated reqs, tests passed

• Added boostrapper test and tests passed

• Adding multi class test script

• Updated Metrics Output

  • Added optl' threshold print inside return_metrics
  • KFold metric printing:
    • Added new input per_fold_print to allow user to return per fold metrics, otherwise average
    • Added tqdm output for KFold metric printing
    • Fixed KFold avg output in report_model_metrics

• Added a regression test and updating report_model_metrics to work with regression and multi class

• Augmented predict_proba test, and train_val_test_split

• Fixed pipeline_steps arg in model definition

• Refactored metrics_df in report_model_metrics for aesthetics

• Unit Tests

  • XGB early stopping multi class test
  • Added fit method tests
  • Added early stopping test
  • Added get_best_score_params() tests
  • Added return_bootstrap_metrics() tests
  • Added tests for get_preprocessing_and_feature_selection_pipeline and get_feature_selection_pipeline
  • Tested init, passed tests
  • Tested get_preprocessing_pipeline

• Imbalance Sampler

  • Addded process_imbalance_sampler() tests, passed
  • Renamed process_imbalance_sampler() to verify_imbalance_sampler

• Made return_dict optional in return_metrics

• Added openpyxl versions for all python versions in requirements.txt

• Refactor metrics, foldwise metrics and foldwise con_mat, class_labels

• Cleaned notebooks dir

• Renamed notebooks to py_example_scripts, linted files, cleaned code

• Added model_tuner version print to scripts

• Added fix for sort of pipelinesteps now optional:

  • Added required model_tuner import to xgb_multi.py
  • Added requisite model_tuner import to multi_class_test.py

• Added catboost_multi_class.py script

• Removed pip dependency from requirements

Model Tuner 0.0.23a

• Fixed a bug found when calibrating early stopping models
• Fixed early stopping in Column Transformer application

Model Tuner 0.0.22a

• Fixed an issue where the feature selection name was not referenced correctly, causing a bug when printing selected feature names with the updated pipeline.
• Removed resolved print statements from April, 2024.

Model Tuner 0.0.21a

• Specified the pipeline class otherwise the method just returned a list
• Removed need to specify self.estimator when its called
• Generalized (renamed) "K Best Features" to just "Best Features" inside returns of return_metrics()
• Generalized (renamed) k_best_features to best_features

Model Tuner 0.0.20a

• Added flexibility between boolean and None for stratification inputs
• Added custom exception for non pandas inputs in return_bootstrap_metrics
• Enforced required model_type input to be specified as "classification" or "regression"
• Removed extraneous "=" print below pipeline_steps
• Handled missing pipeline_steps when using imbalance_sampler
• Updated requirements for python==3.11
• Fixed SMOTE for early stopping
• Removed extra model_type input from xgb_early_test.py

Model Tuner 0.0.19a

• Requirements updated again to make compatible with google colab out of the box.
• Bug in fit() method where best_params wasn't defined if we didn't specify a score
• Threshold bug now actually fixed. Specificity and other metrics should reflect this. (Defaults to 0.5 if optimal_threshold is not specified).

Model Tuner 0.0.18a

• Updated requirements to include numpy versions <1.26 for Python 3.8-3.11.

This should stop a rerun occurring when using the library on a google colab.

Model Tuner 0.0.17a

Major fixes:

• Verbosity variable is now popped from the parameters before the fit
• Bug with Column Transformer early stopping fixed (valid set is now transformed correctly)
• Return metrics now has a consistent naming convention
• report_model_metrics is now using the correct threshold in all cases
• Default values updated for train_val_test_split
• tune_threshold_Fbeta is now called with the correct number of parameters in all cases
• Requirements updates: XGBoost updated to 2.1.2 for later Python versions.

Minor changes:

• help(model_tuner) should now be correctly formatted in google colab

Model Tuner 0.0.16a

Version 0.0.16a

• Custom pipeline steps now updated (our pipeline usage has been completely changed and should now order itself and support non named steps) always ensures correct order
• This fixed multiple other issues that were occuring to do with logging of imbalanced learn
• Reporting model metrics now works.
• AutoKeras code deprecated and removed.
• KFold bug introduced because of CatBoost. This has now been fixed.
• Pretty print of pipeline.
• Boosting variable has been renamed.
• Version constraints have been updated and refactored.
• tune_threshold_Fbeta has been cleaned up to remove unused parameters.
• train_val_test unnecessary self removed and taken outside of class method.
• deprecated setup.py in favor of pyproject.toml per forthcoming pip25 update.

Model Tuner 0.0.15a

Version 0.0.15a

Contains all previous fixes relating to:

• CatBoost support (early stopping, and support involving resetting estimators).
• Pipeline steps now support hyperparameter tuning of the resamplers (SMOTE, ADASYN, etc.).
• Removed older implementations of impute and scaling and moved onto supporting only custom pipeline_steps.
• Fixed bugs in stratification with regards to length mismatch of dependent variable when using column names to stratify.
• Cleaned a removed multiple lines of unused code and unused initialisation parameters.