A partial dependence curve marginalizes the forest's prediction over all other predictors: for each evaluation point of the target variable, the forest scores every training observation with that value substituted in, then averages the result. What you get is the average effect of the target variable after "integrating out" the rest -- a curve that would be flat if the variable carried no signal.
gg_partial_rfsrc(
rf_model,
xvar.names = NULL,
xvar2.name = NULL,
newx = NULL,
partial.time = NULL,
partial.type = c("surv", "chf", "mort"),
cat_limit = 10,
n_eval = 25
)A named list with two elements:
A data.frame with columns x (numeric),
yhat, name (variable name), and optionally grp
(the level of xvar2.name) and time (survival forests
only) for all continuous predictors.
A data.frame with the same columns but
x kept as character, for low-cardinality predictors.
A fitted rfsrc object.
Character vector of predictor names for which partial
dependence should be computed. Must be a subset of rf_model$xvar.names.
Optional single character name of a grouping variable in
newx. When supplied, partial dependence is computed separately for
each unique level of this variable and a grp column is appended.
Optional data.frame of predictor values to evaluate
partial effects at. Defaults to the training data stored in
rf_model$xvar. All column names must match rf_model$xvar.names.
Numeric vector of desired time points for survival
forests (ignored for regression/classification). Values are automatically
snapped to the nearest entry in rf_model$time.interest; see the
Survival forests section below. When NULL (default),
three quartile points of time.interest are used.
Character; type of predicted value for survival
forests, passed through to partial.rfsrc.
One of "surv" (default), "chf", or "mort". Ignored
for non-survival forests. partial.rfsrc() requires a non-NULL
value for survival families; supplying it here avoids a cryptic
“argument is of length zero” error from the underlying C code.
Variables with fewer than cat_limit unique values in
newx are treated as categorical; all others are continuous.
Defaults to 10.
Number of evaluation points for continuous variables. Instead of passing all observed values (which can be slow, especially for survival forests), continuous predictors are evaluated on a quantile grid of this many points. Categorical variables always use all unique levels. Defaults to 25.
partial.rfsrc expects every value in
partial.time to be an exact member of the model's
time.interest vector, the unique observed event times stored in the
fitted object. Pass an arbitrary time, even a plausible one such as
c(1, 3) for a study measured in years, and you get a C-level
prediction error from inside partial.rfsrc.
gg_partial_rfsrc takes care of this: every element of
partial.time is silently snapped to its nearest
time.interest value before the call. To target a specific
follow-up horizon, find the closest grid point yourself and pass it
explicitly:
ti <- rf_model$time.interest
t1 <- ti[which.min(abs(ti - 1))] # nearest to 1 year
pd <- gg_partial_rfsrc(rf_model, xvar.names = "x", partial.time = t1)
partial.rfsrc does not handle
logical predictor columns correctly in survival forests
(randomForestSRC <= 3.5.1). If your training data contains binary 0/1
columns, convert them to factor rather than logical
before fitting the model.
This function builds those curves for one or more predictors by calling
partial.rfsrc and then tidy-stacking the
results into separate data frames for continuous and categorical variables.
Unlike gg_partial (which wraps plot.variable), you
pass the fitted rfsrc object directly -- no intermediate
plot.variable step.
For survival forests, the marginalized quantity depends on
partial.type: survival probability ("surv"), cumulative
hazard function ("chf"), or expected mortality ("mort").
You can request the curve at one or more time horizons via
partial.time; the resulting data have a time column so the
plot layers them as separate coloured lines.
gg_partial, partial.rfsrc,
get.partial.plot.data
## ------------------------------------------------------------
##
## regression
##
## ------------------------------------------------------------
airq.obj <- randomForestSRC::rfsrc(Ozone ~ ., data = airquality)
## partial effect for wind
prt_dta <- gg_partial_rfsrc(airq.obj,
xvar.names = c("Wind"))
Run the code above in your browser using DataLab