Skip to content

FILE: docs/api/seaborn.md

Seaborn API Reference

Functions for exploratory data analysis: correlation heatmaps, distribution plots, and theme setup.

plot_corr(df, method='pearson', threshold=None, figsize='auto', cmap='coolwarm', annot=True, fmt='.2f', show=True, save=None, dpi=150)

Plot a correlation heatmap with a triangular mask.

Parameters: - df – pandas DataFrame (numeric columns only) - method – 'pearson', 'spearman', or 'kendall' - threshold – if not None, highlight pairs with |corr| >= threshold and print list - figsize – tuple or 'auto' (max(8, n_cols*0.8), max(6, n_cols*0.7)) - cmap – colormap for the heatmap - annot – show correlation values in cells - fmt – number format for annotations - show – call plt.show() - save – file path to save figure - dpi – resolution for saved figure

Returns: matplotlib Axes

Example: 

dk.plot_corr(df)
dk.plot_corr(df, method='spearman', threshold=0.5, figsize=(10, 8))

plot_distributions(df, cols=None, kde=True, include_boxplot=True, n_cols=2, figsize='auto', show=True, save_dir=None, dpi=150)

Plot histogram + boxplot for each numeric column.

Parameters: - df – pandas DataFrame - cols – list of column names (None = all numeric columns) - kde – overlay kernel density estimate on histogram - include_boxplot – show boxplot alongside histogram - n_cols – number of panels per row (histogram + boxplot count as separate) - figsize – tuple or 'auto' (default (12, 4) per row) - show – call plt.show() after each figure - save_dir – directory to save individual figures (None = don't save) - dpi – resolution for saved figures

Returns: list of figure objects (one per column)

Example: 

dk.plot_distributions(df)
dk.plot_distributions(df, cols=['age', 'income'], kde=False)
dk.plot_distributions(df, save_dir='figures/', show=False)

set_theme(style='whitegrid', palette='muted', font_scale=1.2, dpi=120, from_config=False, config_path='.datakit.toml')

Set up seaborn and matplotlib theme in one call.

Parameters: - style – seaborn style ('whitegrid', 'darkgrid', 'white', 'dark', 'ticks') - palette – seaborn colour palette - font_scale – multiplier for font sizes - dpi – figure DPI for matplotlib - from_config – load settings from .datakit.toml and override parameters - config_path – path to config file (used if from_config=True)

Returns: None

Example: 

dk.set_theme()
dk.set_theme(style='darkgrid', palette='colorblind', font_scale=1.5)
dk.set_theme(from_config=True)