explain()will now pass
...on to the relevant
gower_powargument to modify the calculated gower distance before use by raising it to the power of the given value (#158)
lime()now warns when quantile binning is not feasible and uses standard binning instead (#154)
lambdavalue in the local model fit to match the one used in the Python version according to the relationship given here: https://stats.stackexchange.com/a/270705
lime.data.frameto keep it in line with the other types. Use it to transform your data.frame into a new input that your model expects after permutations
magickis now only in suggest to cut down on heavy hard dependencies
explainnow returns a
tbl_dfso you get pretty printing if you have
plot_featuresnow has a
casesargument for subsetting the data before plotting
as_regressor()for ad-hoc specification of the model type in case the heuristic implemented in
as_classifier()also lets you add/overwrite the class labels.
goweras the new default similarity measure for tabular data
bin_continuous = FALSEthe default behavior is now to sample from a kernel density estimation rather than assume a normal distribution.
plot_text_explanation()with better formatting and scrolling support for many explanations
NEWS.mdfile to track changes to the package.
POSIXtcolumns. They will be kept constant during permutations so that
limewill explain the model behaviour at the given timepoint based on the remaining features (#39).
plot_explanations()for an overview plot of a large explanation set