anchor_python_visualization.visualize

Different schemes for visualizing embeddings or other types of data.

Package Contents

Classes

VisualizeFeaturesScheme

A method to visualize a set of feature-values.

Functions

create_method(...)

Creates a visualize-embeddings method from an identifier.

Attributes

DEFAULT_IDENTIFIER	The default choice to use in IDENTIFIERS.
IDENTIFIERS	Unique strings to use as command-line-arguments to select a VisualizeFeaturesScheme.

anchor_python_visualization.visualize.DEFAULT_IDENTIFIER = 'plot'

The default choice to use in IDENTIFIERS.

anchor_python_visualization.visualize.IDENTIFIERS = ['plot', 'tensorboard']

Unique strings to use as command-line-arguments to select a VisualizeFeaturesScheme.

All are lower-case.

anchor_python_visualization.visualize.create_method(identifier: str | None, projector: anchor_python_visualization.projection.Projector | None, output_path: str | None) → anchor_python_visualization.visualize.visualize_features_scheme.VisualizeFeaturesScheme

Creates a visualize-embeddings method from an identifier.

Parameters:

• identifier – string that is one of IDENTIFIERS, case-insensitive.

• projector – method for performing projection into smaller dimensionality.

• output_path – a path for writing any relevant output.

Returns:

a newly created instance corresponding to the identifier.

class anchor_python_visualization.visualize.VisualizeFeaturesScheme

Bases: abc.ABC

A method to visualize a set of feature-values.

visualize_data_frame(features: anchor_python_visualization.embeddings.LabelledFeatures) → None

Visualizes the embeddings in some manner.

Parameters:

features – embeddings in a data-frame with associated labels.