Decile analysis#
scikitplot.decile groups model scores into ordered quantiles/ntiles so you can explain response concentration, lift, cumulative gain, KS separation and business value to technical and non-technical stakeholders.
Important
A decile plot only makes sense when the score column and positive-class semantics are correct. Verify pos_label/class_index and score ordering before interpreting lift, gain, or financial curves.
Create a decile table or model plot from labels and predicted scores.
Decile tables, lift, cumulative gain, KS statistic, and reports.
Current top-level ModelPlotPy surface and legacy compatibility namespace.
Response, cumulative lift/gain, costs, profit, ROI, and selection summaries.
Positive-label semantics, legacy behavior, RNG side effects, and current snapshot limitations.
Which page should I use?#
Start with Getting started. Then choose the page that owns the decision you are making: architecture and lifecycle, configuration, persistence, performance, public API, or limits. The guide intentionally separates those concerns so examples do not accidentally become policy.
Examples#
See Decile for executable examples that accompany this guide.