Plot prediction error according to various metrics.
Source:R/visualizer-lib-prediction.R
plot_pred_err.Rd
Plot the raw or summarized prediction errors as a boxplot, scatter plot, line plot, or bar plot with or without 1 SD error bars.
Usage
plot_pred_err(
fit_results,
eval_results = NULL,
evaluator_name = NULL,
vary_params = NULL,
metrics = NULL,
show = c("point", "line"),
...
)
Arguments
- fit_results
A tibble, as returned by the
fit
method.- eval_results
A list of result tibbles, as returned by the
evaluate
method.- evaluator_name
Name of
Evaluator
containing results to plot. To compute the evaluation summary results from scratch or if the evaluation summary results have not yet been evaluated, set toNULL
.- vary_params
A vector of parameter names that are varied across in the
Experiment
.- metrics
A
metric_set
object indicating the metrics to plot. Seeyardstick::metric_set()
for more details. DefaultNULL
will use the default metrics inyardstick::metrics()
.- show
Character vector with elements being one of "boxplot", "point", "line", "bar", "errorbar", "ribbon" indicating what plot layer(s) to construct.
- ...
Additional arguments to pass to
plot_eval_summary()
. This includes arguments for plotting and for passing intosummarize_pred_err()
.
Value
If interactive = TRUE
, returns a plotly
object if
plot_by
is NULL
and a list of plotly
objects if
plot_by
is not NULL
. If interactive = FALSE
, returns
a ggplot
object if plot_by
is NULL
and a list of
ggplot
objects if plot_by
is not NULL
.
See also
Other prediction_error_funs:
eval_pred_curve_funs
,
eval_pred_err_funs
,
plot_pred_curve()
Examples
# generate example fit_results data
fit_results <- tibble::tibble(
.rep = rep(1:2, times = 2),
.dgp_name = c("DGP1", "DGP1", "DGP2", "DGP2"),
.method_name = c("Method"),
# true response
y = lapply(1:4, FUN = function(x) rnorm(100)),
# predicted response
predictions = lapply(1:4, FUN = function(x) rnorm(100))
)
# generate example eval_results data
eval_results <- list(
`Prediction Errors` = summarize_pred_err(
fit_results, truth_col = "y", estimate_col = "predictions"
)
)
# create errorbar plot using pre-computed evaluation results
plt <- plot_pred_err(fit_results = fit_results, eval_results = eval_results,
evaluator_name = "Prediction Errors",
show = c("point", "errorbar"))
# or alternatively, create the same plot without pre-computing evaluation results
plt <- plot_pred_err(fit_results, show = c("point", "errorbar"),
truth_col = "y", estimate_col = "predictions")
# can customize plot (see plot_eval_summary() for possible arguments)
plt <- plot_pred_err(fit_results = fit_results, eval_results = eval_results,
evaluator_name = "Prediction Errors",
show = c("point", "errorbar"),
color_str = NULL,
facet_formula = .method_name ~ .metric,
facet_type = "grid")