The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
calculate_variable_splits() now treats
integer variables as categorical. This change
is propagated to ceteris_paribus(),
partial_dependence(),
accumulated_dependence(),
conditional_dependence(),
aggregate_profiles(),
DALEX::predict_profile(),
DALEX::model_profile()ceteris_paribus /
calculate_variable_splits when tidymodels uses
integer variables #145show_observations #148.
This change is propagated to DALEX::plot.predict_profile()
#540.class(x) = "y" with
is(x, "y")facet_scales parameter to
plot.aggregated_profiles_explainer ('free_x'
by default) #138
and plot.ceteris_paribus_explainer ('free_x'
or 'free_y' by default, depending on plot type) #136N = NULL in
partial_dependence() etc. #134plot.ceteris_paribus_explainer now by default for
categorical variables plots profiles (not lines -prev default- nor
bars)subtitle value in plot.fi changed
to NULL from NA (unification)ceteris_paribus function one can specify how
grid points shall be calculated, see
variable_splits_typeceteris_paribus and aggregates are now working with
missing data, this solves #120plot(ceteris_paribus) change default color
to label or ids if more than one profile is detected,
this solves #123ceteris_paribus has now argument
variable_splits_with_obs which included values from
new_observations in the variable_splits, this
solves #124n_sample argument in
feature_importance (now it’s N) #113plot_profile now handles multilabel modelsDALEX is moved to Suggests as in #112plot_categorical_ceteris_paribus can plot bars
(again)bind_plots functionR v4.0 and depend on R v3.5 to
comply with DALEXtitle and subtitle in
several plotsdependency to dependence #103ceteris_paribus profiles are now working for
categorical variablesshow_profiles, show_observations,
show_residuals are now working for categorical
variablesdesc_sorting in
plot.variable_importance_explainer #94feature_importance now does 15
permutations on each variable by default. Use the B
argument to change this numberplot.feature_importance and
plotD3.feature_importance that showcase the permutation
dataaggregate_profiles: preserve _x_ column
factor order and sort its values #82aggregate_profiles use now gaussian kernel smoothing.
Use the span argument for fine control over this parameter
(#79)variable_type and variables
arguments usage in the aggregate_profiles,
plot.ceteris_paribus and
plotD3.ceteris_paribusvariable_type argument from
plotD3.aggregated_profiles (now the same as in
plot.aggregated_profiles)DALEXtra as
aspect_importance is moved to DALEXtra as well
(See
v0.3.12 changelog)aspect_importance is moved to DALEXtra (#66)titanic_imputed from DALEX (#65)plot.aspect_importance - it can plot more than
single figuretriplot, plot.aspect_importance
and plot_group_variables to add more clarity in plots and
allow some parameterizationtriplot function that illustrates hierarchical
aspect_importance() groupingsaspect_importance() functionsaspect_importance()only_numerical parameter to
variable_type in functions aggregated_profiles(),
cluster_profiles(), plot() and others, as requested in #15describe()
function for ceteris_paribus(),
feature_importance() and aggregate_profiles()
explanations.aggregated_profiles_conditional and
aggregated_profiles_accumulated are rewritten with some
code fixeslime is implemented in the
lime()/aspect_importance() function.B that replicates
permutations B times and calculates average from drop
loss.plotD3 now supports Ceteris Paribus Profiles.feature_importance now can take
variable_grouping argument that assess importance of group
of featuresshow_profiles and show_residuals functions
extend Ceteris Paribus Plots.show_aggreagated_profiles is renamed to
show_aggregated_profilesdescribe() and
print.ceteris_paribus_descriptions() for text based
descriptions of Ceteris Paribus explainersplot.ceteris_paribus_explainer works now also for
categorical variables. Use the only_numerical = FALSE to
force barspartial_profiles(), accumulated_profiles()
and conditional_profiles for variable effectsceteris_paribus_2d extends classical ceteris paribus
profilesceteris_paribus_oscillations calculates oscilations for
ceteris paribus profilescluster_profiles helps to identify interactionspartial_dependency calculates partial dependency
plotsaggregate_profiles calculates partial dependency plots
and much moremodel_feature_importance and
model_feature_response from DALEX to
ingredientsThese binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.