plot.UNCOVER {UNCOVER}

Plot various outputs of UNCOVER

Description

Allows visualisation of many aspects of UNCOVER, including co-variate, posterior and diagnostic plots.

Usage

## S3 method for class 'UNCOVER'
plot(
  x,
  type = "covariates",
  plot_var = x$Minimum_Spanning_Tree_Variables,
  diagnostic_x_axis = "full",
  ...
)

Arguments

x

Object of class "UNCOVER"

type

Can be one of; "covariates" for cluster assignment visualisation for the co-variates, "fitted" for co-variate visualisation with respect to their fitted values, "samples" for posterior visualisation or "diagnostics" for diagnostic plots. See details. Defaults to "covariates".

plot_var

Vector specifying which columns (or associated logistic regression coefficients) of the co-variate matrix should be plotted. Does not apply when type=="diagnostics". Defaults to all columns being selected.

diagnostic_x_axis

Only applies if ⁠"type=="diagnostics"⁠. Either "full" (default) for all observations indices to be plotted on the x-axis or "minimal" for only some observations indices to be plotted on the x-axis.

...

Arguments to be passed to methods

Details

If type=="covariates", the resulting plot will be a ggpairs plot. The diagonal entries will be a plot of K density plots (K being the number of clusters). The off-diagonal elements are scatter-plots of the observations, given a label according to their true response and a colour based on their assigned cluster. If length(plot_var)==1 then the density plot and the scatter-plot are combined. If a cluster contains less than two data points the density will not be plotted.

If ⁠"type=="fitted"⁠, the resulting plot will be a ggpairs plot. The diagonal entries will be two density plots, one for data predicted to have response 0 by the model (red) and one for training data predicted to have response 1 by the model (green). The off-diagonal elements are scatter-plots of the observations, given a label according to their actual response and a colour scale based on their predicted response. If length(plot_var)==1 then the density plot and the scatter-plot are combined. If a predicted class (0 or 1) contains less than two data points the density will not be plotted.

If type=="samples", the resulting plot will be a ggpairs plot of the clusters posteriors, giving the coefficient densities in the diagonal and scatter-plots of the posterior samples in the off-diagonal. The transparency is increased in the upper triangle for scenarios when posteriors overlap.

If "type==diagnostics", the resulting plot depends on the deforestation criterion used to create the "UNCOVER" object:

"None"

A plot tracking the overall log Bayesian evidence every time an action is executed.

"NoC"

A plot tracking the overall log Bayesian evidence after every action and a plot tracking the number of clusters after every action.

"SoC"

Three plots; one tracking the overall log Bayesian evidence after every action, one tracking the number of criterion breaking clusters after every action and one tracking the minimum cluster size after every action.

"MaxReg"

A plot tracking the overall log Bayesian evidence every time an action is executed. Actions are coloured and each action has an associated coloured dashed line indicating the log Bayesian evidence plus the logarithm of the maximum tolerance provided.

"Validation"

A plot tracking the overall log Bayesian evidence after every action (for both the training data and all of the data) and a plot tracking the robustness statistic after every deforestation action.

"Diverse"

Three plots; one tracking the overall log Bayesian evidence after every action, one tracking the number of criterion breaking clusters after every action and one tracking the minimum minority class across clusters after every action.

Actions are defined as either edge removals, edge additions or edge additions in the deforestation stage. The syntax for an action will be the 'type_of_action.edge'. For example the removal of an edge connecting observation 1 and observation 2 will be displayed 'Rem.1-2'. If the edge was being added this would be displayed 'Def.Add.1-2' if in the deforestation stage and 'Add.1-2' otherwise. When the data for the "UNCOVER" object created is large setting diagnostic_x_axis=="minimal" is recommended as it gives a more visually appealing output.

Value

No return value, called for side effects

See Also

UNCOVER()

Examples

require(graphics)
# First we generate a co-variate matrix and binary response vector
CM <- matrix(rnorm(200),100,2)
rv <- sample(0:1,100,replace=TRUE)

# We can then run our algorithm for each of the different deforestation
# criteria
UN.none <- UNCOVER(X = CM,y = rv, deforest_criterion = "None", verbose = FALSE)
UN.noc <- UNCOVER(X = CM,y = rv, deforest_criterion = "NoC",
                  options = UNCOVER.opts(max_K = 3), verbose = FALSE)
UN.soc <- UNCOVER(X = CM,y = rv, deforest_criterion = "SoC",
                  options = UNCOVER.opts(min_size = 10), verbose = FALSE)
UN.maxreg <- UNCOVER(X = CM,y = rv, deforest_criterion = "MaxReg",
                     options = UNCOVER.opts(reg = 1), verbose = FALSE)
UN.validation <- UNCOVER(X = CM,y = rv, deforest_criterion = "Validation",
                         options = UNCOVER.opts(train_frac = 0.8),
                         verbose = FALSE)
UN.diverse <- UNCOVER(X = CM,y = rv, deforest_criterion = "Diverse",
                       options = UNCOVER.opts(n_min_class = 2), verbose = FALSE)
plot(UN.none,type = "covariates")
plot(UN.none,type = "fitted")
plot(UN.none,type = "samples")
plot(UN.none,type = "diagnostics",diagnostic_x_axis = "minimal")
plot(UN.noc,type = "diagnostics",diagnostic_x_axis = "minimal")
plot(UN.soc,type = "diagnostics",diagnostic_x_axis = "minimal")
plot(UN.maxreg,type = "diagnostics",diagnostic_x_axis = "minimal")
plot(UN.validation,type = "diagnostics",diagnostic_x_axis = "minimal")
plot(UN.diverse,type = "diagnostics",diagnostic_x_axis = "minimal")

# If we only wanted to view the second co-variate
plot(UN.none,type = "covariates",plot_var=2)
plot(UN.none,type = "fitted",plot_var=2)
plot(UN.none,type = "samples",plot_var=2)