break_down
function identifies inteactionsbreak_down
function supports DALEX explainersbreak_down
function has complexity O(2p) for models without interactions, much faster than the old version
broken.default
has now thekeep_distributions
arguments. IfTRUE
then the whole distribution of conditional residuals is remebered and avaliable for plotting #17- small updates in
README.md
- small changes in
broken.default
to make it work withxgboost
and other nondata.frame
data broken.lm
supports unnormalized coefficients (thanks to Joseph Larmarange) just addpredict.function = betas
#9
broken.default
is now model agnostic!broken.ranger
is removed sincebroken.default
is much better- small fixes in
print
andplot
functions, a new vigniette for model agnostic plots
- small fixes and submission to CRAN
broken.lm
andbroken.glm
are now supporting interactions (#7)print()
andplot()
functions are now handling different options for rounding via additional argumentsdigits = 3
,rounding_function = round
(#8)
- the
baseline
argument is added to thebroken
function (#1) - vignettes for lm and glm models are added (#2)
- waterfall like plots and support for lm models
- waterfall like plots and support for glm models
- HR dataset added