plot_umap

obsidian.experiment.advanced_design.plot_umap(design, continuous_params_keys, subparam_mapping, hue=None, n_neighbors=15, min_dist=0.1, metric='euclidean', verbose=False)

Plot a UMAP projection of the continuous and subparameter columns.

Requires the umap-learn package (pip install umap-learn).

Parameters:

• design – Design DataFrame.

• continuous_params_keys – List of continuous parameter column names.

• subparam_mapping – Dict mapping categorical variables to their subparameters.

• hue – Column name to use for color-coding. Defaults to None.

• n_neighbors – UMAP neighborhood size. Defaults to 15.

• min_dist – UMAP minimum distance parameter. Defaults to 0.1.

• metric – Distance metric for UMAP. Defaults to 'euclidean'.

• verbose – Whether to show UMAP’s internal progress log. Defaults to False.

Raises:

ImportError – If umap-learn is not installed.