Extension for 'DALEX' Package
Compare machine learning models
Create your conda virtual env with DALEX
DALEX load explainer
Create explainer from your h2o model
Wrapper for Python Keras Models
Create explainer from your mlr model
Create explainer from your mlr model
Wrapper for Python Scikit-Learn Models
Create explainer from your tidymodels workflow.
Create explainer from your xgboost model
Caluculate difference in performance in models across different catego...
Exract info from model
Compare champion with challengers globally
Funnel plot for difference in measures
Plot function for overall_comparison
Plot and compare performance of model between training and test set
Instance Level Surrogate Models
Print funnel_measure object
Print overall_comparison object
Prints scikitlearn_set class
Print funnel_measure object
Compare performance of model between training and test set
Wrapper over the predict function
Provides wrapper of various machine learning models. In applied machine learning, there is a strong belief that we need to strike a balance between interpretability and accuracy. However, in field of the interpretable machine learning, there are more and more new ideas for explaining black-box models, that are implemented in 'R'. 'DALEXtra' creates 'DALEX' Biecek (2018) <doi:10.48550/arXiv.1806.08915> explainer for many type of models including those created using 'python' 'scikit-learn' and 'keras' libraries, and 'java' 'h2o' library. Important part of the package is Champion-Challenger analysis and innovative approach to model performance across subsets of test data presented in Funnel Plot.
Useful links