Plot a PCA Fit — pca_plot • TempleCBE

One entry point to this package's PCA plots.

Usage

pca_plot(
  pca_model,
  type = c("variance", "heatmap", "bi", "biplot", "scree", "circle"),
  ...
)

Arguments

pca_model: A prcomp object or cbe_pca object from run_pca. (Not `x`, which would capture the `x` component argument of `type = "bi"` and `"biplot"`.)
type: One of "variance" (pca_percent_var_explained), "heatmap" (pca_feature_loading_heatmap), "bi" (plot_pca_bi), "biplot" (pca_biplot), "scree" (pca_scree_plot), or "circle" (pca_variables_plot).
...: Passed on to the underlying plot function (needed for type = "bi", which requires newdata and column; and optionally used by type = "bi" or "biplot" to pick components with x/y).

Value

A ggplot object.

Details

This is an ordinary function rather than a `plot()` method: stats already registers `plot()` for `prcomp` objects (a scree plot), and a package that replaces another package's method for a class it doesn't own changes `plot()` for everyone who loads it. Call `pca_plot(x, type = )`, or [stats::screeplot()] for the base scree plot.

Examples

pca_model <- prcomp(mtcars, center = TRUE, scale. = TRUE)
pca_plot(pca_model, type = "variance")

pca_plot(pca_model, type = "biplot", x = 1, y = 3)