Contents
visualize$evaluate$redundancy
Model redundancy heatmap
Model-by-model geom_tile of one evaluate$redundancy() method.
The upper-triangle pairs the result rows carry are mirrored across the
diagonal so every model pair is drawn once in each direction; the diagonal
itself is left blank, since no self-pair is computed.
Usage
visualize.evaluate.redundancy(
results,
method = NULL,
theme = c("polygenius", "none")
)Arguments
| Argument | Description |
|---|---|
results | A [PolyGeniusEvaluation](/reference/evaluate/) of evaluate$redundancy(). |
method | Character scalar naming the method to plot, or NULL (default), which takes the first method present in metric-table order (score.pearson, snp.jaccard, snp.weighted.overlap). An unknown value aborts, naming the methods results actually carries. |
theme | One of "polygenius" (default), "none". Plot theme. "none" gives a bare theme_minimal() to style yourself; palette colors are applied either way. |
Value
A ggplot. score.pearson, a correlation, uses the diverging
palette on [-1, 1]; snp.jaccard and snp.weighted.overlap use the
sequential palette on [0, 1].
Details
The grid is models x models: its tile count grows with the square of the model count, with no bound. A large library renders a large, slow tile.
Examples
red <- evaluate$redundancy(data)
visualize$evaluate$redundancy(red, method = "score.pearson")See Also
evaluate$redundancy(), which produces results.
Other visualize-evaluate:
visualize.evaluate.association(),
visualize.evaluate.compare(),
visualize.evaluate.incremental(),
visualize.evaluate.performance(),
visualize.evaluate.profile(),
visualize.evaluate.stratification()