plot_R_NX {dimRed}

plot_R_NX

Description

Plot the R_NX curve for different embeddings. Takes a list of dimRedResult objects as input. Also the Area under the curve values are computed for a weighted K (see AUC_lnK_R_NX for details) and appear in the legend.

Usage

plot_R_NX(x, ndim = NA, weight = "inv")

Arguments

x

a list of dimRedResult objects. The names of the list will appear in the legend with the AUC_lnK value.

ndim

the number of dimensions, if NA the original number of embedding dimensions is used, can be a vector giving the embedding dimensionality for each single list element of x.

weight

the weight function used for K when calculating the AUC, one of c("inv", "log", "log10")

Value

A ggplot object, the design can be changed by appending theme(...)

See Also

Other Quality scores for dimensionality reduction: AUC_lnK_R_NX,dimRedResult-method, LCMC,dimRedResult-method, Q_NX,dimRedResult-method, Q_global,dimRedResult-method, Q_local,dimRedResult-method, R_NX,dimRedResult-method, cophenetic_correlation,dimRedResult-method, distance_correlation,dimRedResult-method, mean_R_NX,dimRedResult-method, quality,dimRedResult-method, reconstruction_error,dimRedResult-method, reconstruction_rmse,dimRedResult-method, total_correlation,dimRedResult-method

Examples

if(requireNamespace(c("RSpectra", "igraph", "RANN", "ggplot", "tidyr", "scales"), quietly = TRUE)) {
## define which methods to apply
embed_methods <- c("Isomap", "PCA")
## load test data set
data_set <- loadDataSet("3D S Curve", n = 200)
## apply dimensionality reduction
data_emb <- lapply(embed_methods, function(x) embed(data_set, x))
names(data_emb) <- embed_methods
## plot the R_NX curves:
plot_R_NX(data_emb) +
    ggplot2::theme(legend.title = ggplot2::element_blank(),
                   legend.position = c(0.5, 0.1),
                   legend.justification = c(0.5, 0.1))
}