UMAP embedding for waveform data

Performs UMAP dimensionality reduction on waveform data for visualization. Requires the uwot package.

Usage

waveformUMAP(
  x,
  n_neighbors = 15,
  n_components = 2,
  min_dist = 0.1,
  metric = "euclidean",
  features = c("statistical", "shape"),
  use_pca = TRUE,
  n_pca = 30,
  seed = NULL
)

Arguments

x

A PhysioExperiment object, matrix, or waveform_pca result.

n_neighbors

Number of neighbors for UMAP.

n_components

Number of UMAP dimensions (usually 2).

min_dist

Minimum distance parameter.

metric

Distance metric: "euclidean", "cosine", "manhattan".

features

If x is waveform data, features to extract.

use_pca

Logical; pre-reduce with PCA (recommended for high-dim).

n_pca

Number of PCA components to use as input.

seed

Random seed for reproducibility.

Value

A list of class "waveform_umap" containing:

embedding

UMAP coordinates (observations x n_components)

n_obs

Number of observations

params

UMAP parameters used

References

van der Maaten L, Hinton G (2008). "Visualizing Data using t-SNE." Journal of Machine Learning Research, 9, 2579-2605.

See also

waveformPCA(), waveformTSNE(), plotUMAP(), extractWaveformFeatures()

Examples

if (FALSE) { # \dontrun{
# UMAP on gait data
data <- matrix(rnorm(1000), nrow = 100, ncol = 10)
umap_result <- waveformUMAP(data, n_neighbors = 15)
plotUMAP(umap_result)
} # }