Extra margin to pass to ylim as a fraction of the range of x$ice_curves.
frac_to_plot
If frac_to_plot is less than 1, randomly plot frac_to_plot fraction of the
curves in x$ice_curves.
plot_points_indices
If not NULL, this plots only the indices of interest. If not NULL, frac_to_plot must be 1 otherwise
an error is thrown. Default is NULL.
plot_orig_pts_preds
If TRUE, marks each curve at the location of the observation's actual fitted value. If FALSE,
no mark is drawn.
pts_preds_size
Size of points to make if plot_origin_pts_preds is TRUE.
colorvec
Optional vector of colors to use for each curve.
color_by
Optional variable name in Xice, column number in Xice, or data vector of the correct length to color curves by.
If the color_by
variable has 10 or fewer unique values, a discrete set of colors is used for each value and a legend is
printed and returned. If there are more values, curves are colored from light to dark corresponding
to low to high values of the variable specified by color_by.
x_quantile
If TRUE, the plot is drawn with the x-axis taken to be quantile(gridpts). If FALSE,
the predictor's original scale is used.
plot_pdp
If TRUE, the PDP is plotted and highlighted in yellow.
centered
If TRUE, all curves are re-centered to be 0 at the quantile given by centered_percentile.
See Goldstein et al (2013) for details and examples. If FALSE, the original ice_curves are plotted.
prop_range_y
When TRUE and centered=TRUE as well, the range of the right vertical axis displays the
centered values as a fraction of the sd of the fitted values on actual observations if prop_type
is missing or set to "sd". If prop_type is set to "range", the right axis displays the
centered values as a fraction of the range of the fitted values over the actual observations.
centered_percentile
The percentile of predictor for which all ice_curves are "pinched together" and set to be 0.
Default is .01.
point_labels
If not NULL, labels to plot next to each point. Default is NULL.
point_labels_size
If not NULL, size of labels to plot next to each point. Default is NULL which means it's the size of pts_preds_size.
rug_quantile
If not NULL, tick marks are drawn on the x-axis corresponding to the vector of quantiles specified by this parameter.
Forced to NULL when x_quantile is set to TRUE.
prop_type
Scaling factor for the right vertical axis in centered plots if prop_range_y is TRUE. Can be one of
"sd" (default) or "range". Ignored if centered and prop_range_y are not both TRUE.
...
Other arguments to be passed to the plot function.
Value
A list with the following elements.
plot_points_indices
Row numbers of Xice of those observations presented in the plot.
legend_text
If the color_by argument was used,
a legend describing the map between the color_by predictor
and curve colors.
See Also
ice
Examples
## Not run:
require(ICEbox)
require(randomForest)
require(MASS) #has Boston Housing data, Pima
data(Boston) #Boston Housing data
X = Boston
y = X$medv
X$medv = NULL
## build a RF:
bhd_rf_mod = randomForest(X, y)
## Create an 'ice' object for the predictor "age":
bhd.ice = ice(object = bhd_rf_mod, X = X, y = y, predictor = "age",
frac_to_build = .1)
## plot
plot(bhd.ice, x_quantile = TRUE, plot_pdp = TRUE, frac_to_plot = 1)
## centered plot
plot(bhd.ice, x_quantile = TRUE, plot_pdp = TRUE, frac_to_plot = 1,
centered = TRUE)
## color the curves by high and low values of 'rm'.
# First create an indicator variable which is 1 if the number of
# rooms is greater than the median:
median_rm = median(X$rm)
bhd.ice$Xice$I_rm = ifelse(bhd.ice$Xice$rm > median_rm, 1, 0)
plot(bhd.ice, frac_to_plot = 1, centered = TRUE, prop_range_y = TRUE,
x_quantile = T, plot_orig_pts_preds = T, color_by = "I_rm")
bhd.ice = ice(object = bhd_rf_mod, X = X, y = y, predictor = "age",
frac_to_build = 1)
plot(bhd.ice, frac_to_plot = 1, centered = TRUE, prop_range_y = TRUE,
x_quantile = T, plot_orig_pts_preds = T, color_by = y)
## End(Not run)