scatcluster.analysis.dendrogram

Dendrogram analysis module.

Classes

Dendrogram

Functions

get_leaves(dendrogram_info, ax)	Get dendrogram list of leaves coordinates and colors.
get_prediction(linkage, population_size)	Get cluster predection for each sample.
show_dendrogram(linkage[, ax, depth])	Generate a dendrogram plot and return the cluster predictions.
dendrogram(linkage, times, n_clusters[, n_cal_bins])	Generates a dendrogram plot and visualizes cluster properties for a given linkage matrix and times.

Module Contents

scatcluster.analysis.dendrogram.get_leaves(dendrogram_info, ax)

Get dendrogram list of leaves coordinates and colors.

Parameters:

• dendrogram_info (dict) – Output of the scipy.hierarchy.dendrogram function.

• ax (plt.Axes) – The axes to draw the dendrogram into.

Returns:

• leaves_coordinates (array-like) – The x coordinates of each leave.

• leaves_population_size (array-like) – The population size of each leave.

scatcluster.analysis.dendrogram.get_prediction(linkage, population_size)

Get cluster predection for each sample.

The dendrogram (or linkage) matrix sorts the samples per catergory. We revert that back to their correct time coordinate. Note: the clusters are labeled from 1 to n_clusters.

Parameters:

• linkage (np.ndarray) – Output of the linkage function.

• population_size (list) – The size of every cluster

scatcluster.analysis.dendrogram.show_dendrogram(linkage, ax=None, depth=30)

Generate a dendrogram plot and return the cluster predictions.

Parameters:

• linkage (np.ndarray) – The linkage matrix.

• ax (matplotlib.axes.Axes, optional) – The axes to plot the dendrogram on. Defaults to None.

• depth (int, optional) – The depth of the dendrogram. Defaults to 30.

Returns:

The cluster predictions for each sample.

Return type:

np.ndarray

scatcluster.analysis.dendrogram.dendrogram(linkage, times, n_clusters, n_cal_bins=150)

Generates a dendrogram plot and visualizes cluster properties for a given linkage matrix and times.

Parameters:

• linkage (numpy.ndarray) – The linkage matrix representing the hierarchical clustering.

• times (numpy.ndarray) – The array of times associated with each data point.

• n_clusters (int) – The number of clusters to show in the dendrogram.

• n_cal_bins (int, optional) – The number of calendar bins to use for calendar occurrences. Defaults to 150.

Returns:

A tuple containing the figure object and the predictions made by the dendrogram.

Return type:

tuple

class scatcluster.analysis.dendrogram.Dendrogram

process_dendrogram_linkage(**kwargs) → None

Process the full linkage for the Dendrogram. This is managed via dendrogram_method. Any additional parameters apart from ‘method’ can be supplied via kwargs

preload_linkage() → None

Load the linkage for the Dendrogram.

This function loads the linkage for the Dendrogram from a file. The file path is constructed using the attributes data_network, data_station, data_location, network_name, and ica_number_components of the instance. The file path is constructed using the data_savepath attribute of the instance.

single_dendrogram(n_clusters: int, print_predictions: bool = True) → numpy.ndarray | None

Visualise a single Dendrogram based on n_cluster

Parameters:

• n_clusters (int) – The number of clusters to be identifed

• print_predictions (bool, optional) – Whether to print the predictions. Defaults to True.

Returns:

An array with cluster defintion of all the data

Return type:

np.ndarray

range_dendrograms(dendrogram_start: int = 4, dendrogram_end: int = 20) → None

Visualise a range of Dendrograms based on dendrogram_start and dendrogram_end

Parameters:

• dendrogram_start (int, optional) – Defaults to 4.

• dendrogram_end (int, optional) – Defaults to 20.

preload_dendrogram_linkage(n_clusters: int) → None

Preload Dendrogram linkage based on n_clusters.

Parameters:

n_clusters (int) – Number of clusters

_set_share_axes(axs, target=None, sharex=False, sharey=False) → None

Sets the axes sharing properties for a given set of axes.

Parameters:

• axs (numpy.ndarray) – The array of axes to set the sharing properties for.

• target (matplotlib.axes.Axes, optional) – The target axes to share with. If not provided, the first axes in the array will be used.

• sharex (bool, optional) – Whether to share the x-axis between the axes. Defaults to False.

• sharey (bool, optional) – Whether to share the y-axis between the axes. Defaults to False.

show_dendrogram_waveforms(linkage: numpy.array, ax: matplotlib.axes.Axes, depth: int = 30)

Generate a dendrogram plot of waveforms and return the leaf node coordinates and population labels.

Parameters:

• linkage (np.array) – The linkage matrix.

• ax (matplotlib.axes.Axes) – The axes to plot the dendrogram on.

• depth (int, optional) – The depth of the dendrogram. Defaults to 30.

show_dendrogram_with_waveforms(n_clusters: int, experiment_name: str = '')

Generates a dendrogram with waveforms for clustering analysis.

Parameters:

• n_clusters (int) – The number of clusters to generate.

• experiment_name (str, optional) – The name of the experiment. Defaults to an empty string.