depictr 0.2.2

The calibration example fits a logistic regression and plots its predicted probabilities. A reliability curve compares predicted probability with observed frequency, so the previous hand-written score misstated the calibration it was meant to demonstrate.

The seasonal example carries trend and noise. Its series was a noiseless sine, so all ten cycles coincided exactly and the figure showed a single curve behind a ten-entry legend.

calibration_plot gained the Parameters and Returns sections it lacked, including a note that y_score must hold fitted probabilities.

depictr 0.2.1

Corrects the palette_safety() result shape shown in the README: it also returns worst_condition and worst_pair.

The gallery now covers every plot family, matching the R vignettes: interactions, model comparison, VIF, binned residuals, random effects, posteriors, power curves, the model report, precision-recall/gains/lift, the scatter-plot matrix, ridgeline, dumbbell, the summary table, scree/dendrogram/silhouette, autocorrelation, seasonality and the colour-vision simulator.

See CHANGELOG.md.

depictr 0.2.0

Fixes a draw-time crash on missing group values, corrects the optional-extra install hints, raises the plotnine floor to 0.15 for the composition operators, and warns when a figure-level title cannot apply to a multi-panel grid. Full details in CHANGELOG.md.

depictr-py 0.1.1

• Opt-in inside-panel legends via a legend_inside parameter and the legend_inside() theme helper.
• Fixes: invisible grouped histograms, axis/legend titles leaking raw column names, and a duplicated colour scale on survival_plot(risk_table=True).
• Corrected reference titles and added DOIs.

See CHANGELOG.md for detail.