Unverified black box model is the path to the failure. Opaqueness leads to distrust.
Distrust leads to ignoration. Ignoration leads to rejection.
DALEX package xrays any model and helps to explore and explain its behaviour.
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 direct interpretability.
DALEX package contains various methods that help to understand the link between input variables
and model output. Implemented methods help to explore model on the level of a single instance
as well as a level of the whole dataset.
All model explainers are model agnostic and can be compared across different models.
DALEX package is the cornerstone for 'DrWhy.AI' universe of packages for visual model exploration.
Find more details in (Biecek 2018)
apartments_test
and HR_test
yhat
for ranger
and svm
models.single_prediction()
are now consistent with breakDown::broken()
. Specifically, baseline
is now 0
by default instead of "Intercept"
. The user can also specify the baseline
and other arguments by passing them to single_prediction
(@kevinykuo, #39). WARNING: Change in the default value of baseline
.yhat.*
functions help to handle additional parameters to different predict()
functions.CITATION
infoHR
and HRTest
. Target variable is a factor with three levels. Is used in examples for classification.plot.model_performance()
has now show_outliers
parameter. Set it to anything >0 and observations with largest residuals will be presented in the plot. (#34)variable_response()
to better support of gbm
models (c8393120ffb05e2f3c70b0143c4e92dc91f6c823).plot_model_performance()
(e5e61d0398459b78ea38ccc980c4040fd853f449).breakDown
v 0.1.6.single_variable() / variable_response()
function uses predict_function
from explainer
(#17)explain()
function converts tibbles
to data.frame
when specified as data
argument (#15)explain.default()
should help when explain()
from dplyr
is loaded after DALEX
(#16)model_performance()
, variable_importance()
, variable_response()
, outlier_detection()
, prediction_breakdown()
. Old names are now deprecated but still working. (#12)apartments
- will be used in examplesvariable_importance()
allows work on full dataset if n_sample
is negativeplot_model_performance()
uses ecdf or boxplots (depending on geom
parameter).single_variable()
supports factor variables as well (with the use of factorMerger
package). Remember to use type='factor'
when playing with factors. (#10)explain()
. Old version has an argument predict.function
, now it's predict_function
. New name is more consistent with other arguments. (#7)xgboost
model (#11)variable_dropout()
functionexplain()
function implementedsingle_prediction()
function implementedsingle_variable()
function implemented