Fourth release on CRAN. This version introduces significant memory efficiency improvements for large-scale data analysis.
save.memory
option in interpret().max.ncol argument with max.nelements in
interpret() to provide more intuitive control over the
memory consumption of the design matrix.interpret() and
predict.mid().plot.mid.breakdown() and
ggmid.mid.breakdown() now support format.args
and enhanced label.format for better visualization
control.stats::terms()) and improved documentation clarity.interpret() to improve space and time
complexity of constructing the design matrix.format argument in mid.breakdown() is
deprecated.plot.mid.breakdown() and
ggmid.mid.breakdown() now have a new argument
format.args, which is passed to base::format()
to format the predictor values stored in “mid.breakdown” objects.format argument in
plot.mid.breakdown() and ggmid.mid.breakdown()
is renamed to label.format. The formatting strings now
support more flexible formats, such as “%t=%v, %t=%v” for
interactions.ggmid.mid.importance() and
plot.mid.importance() to modify appearance of the plots
when color themes are applied.interpret(): The model object no
longer stores the massive fitted.matrix (the term-wise
decomposition of the fitted values).predict() engine: Re-implemented the
prediction logic using a matrix-free approach.mid.importance() now perform term-wise decomposition
on-the-fly using the new optimized prediction engine.mid.importance() introduced a new argument
max.nkeeps (default: 10,000). While importance scores are
calculated using the full dataset for maximum accuracy, the function now
optionally retains only a weighted random sample of the term-wise
predictions.predict outputs: For
type = "terms", the intercept is now stored in the
constant attribute of the returned matrix, aligning with
standard R conventions (e.g., predict.lm).fitted.matrix reference in
interpret.default to prevent memory leaks during the
estimation process.interpret.formula() now supports unevaluated column
names for the weights argument.weighted.loss() supports the R-squared metrics by
passing method = "r2".NA values.stats::terms() function.Third release on CRAN.
max.nterms, max.nplots,
max.nrow).Second release on CRAN.
factor.encoder() and numeric.encoder() for
improved performance.weighted.rmse() and its related functions into
a single, more versatile weighted.loss() function.max.bars to max.terms, max.nrow
to max.rows, etc.).weighted() and its family functions.mid.extract() and
mid.frames().color.theme() to significantly enhance its
functionality and flexibility.numeric.encoder() and
factor.encoder() held an unnecessary reference to the
execution environment of interpret.default().interpret.formula() and
factor.encoder() to correctly support subset
and drop.unused.levels arguments.get.yhat() methods to ensure prediction outputs
always have the same length as the number of input observations.interpret.default().interpret.default() that caused
inconsistency between “fitted.values” and “residuals”.mid.f() (mid.effect())
to correctly handle vector recycling when an input’s length is 1.autoplot.mid.conditional() to avoid redundant
evaluation of the “mid” object.k) in interpret().color.theme() for
easier theme specification.interpret.formula() to resolve environment
issues related to stats::model.frame()color.theme().midr.diverging, midr.qualitative and
midr.sequential.interpret.formula() to ensure the
evaluated formula is correctly stored in the function
call.First release on CRAN.