DALEX v0.3.0


Monthly downloads



Descriptive mAchine Learning EXplanations

Machine Learning (ML) models are widely used and have various applications in classification or regression. Models created with boosting, bagging, stacking or similar techniques are often used due to their high performance, but such black-box models usually lack of interpretability. DALEX package contains various explainers that help to understand the link between input variables and model output. The single_variable() explainer extracts conditional response of a model as a function of a single selected variable. It is a wrapper over packages 'pdp' (Greenwell 2017) <doi:10.32614/RJ-2017-016>, 'ALEPlot' (Apley 2018) <arXiv:1612.08468> and 'factorMerger' (Sitko and Biecek 2017) <arXiv:1709.04412>. The single_prediction() explainer attributes parts of a model prediction to particular variables used in the model. It is a wrapper over 'breakDown' package (Staniak and Biecek 2018) <doi:10.32614/RJ-2018-072>. The variable_dropout() explainer calculates variable importance scores based on variable shuffling (Fisher at al. 2018) <arXiv:1801.01489>. All these explainers can be plotted with generic plot() function and compared across different models. 'DALEX' is a part of the 'DrWhy.AI' universe (Biecek 2018) <arXiv:1806.08915>.

Functions in DALEX

Name Description
apartments Apartments Data
prediction_breakdown Explanations for a Single Prediction
print.explainer Prints Explainer Summary
theme_drwhy DrWhy Theme for ggplot objects
print.model_performance_explainer Model Performance Summary
yhat Wrapper over the predict function
theme_mi2 MI^2 Theme
titanic Passengers and Crew on the RMS Titanic
plot.variable_response_explainer Plots Marginal Model Explanations (Single Variable Responses)
predict.explainer Wrapper over the predict function
plot.prediction_breakdown_explainer Plots Local Explanations (Single Prediction)
plot.variable_importance_explainer Plots Global Model Explanations (Variable Importance)
variable_importance Loss from Variable Dropout
variable_response Marginal Response for a Single Variable
HR Human Resources Data
install_dependencies Install all dependencies for the DALEX package
loss_cross_entropy Preimplemented Loss Functions
dragons Dragon Data
explain.default Create Model Explainer
model_performance Model Performance Plots
model_feature_response Marginal Response for a Single Variable
plot.model_feature_response_explainer Plots Marginal Model Explanations (Single Variable Responses)
plot.model_performance_explainer Model Performance Plots
No Results!

Last month downloads


License GPL
Encoding UTF-8
LazyData true
RoxygenNote 6.1.1
URL https://pbiecek.github.io/DALEX/
BugReports https://github.com/pbiecek/DALEX/issues
NeedsCompilation no
Packaged 2019-03-25 21:50:43 UTC; pbiecek
Repository CRAN
Date/Publication 2019-03-25 22:46:17 UTC

Include our badge in your README