Partial dependence plot

Q: How deeply does Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI cover partial dependence plot?

Defines averaging and independence assumptions, demonstrates plots, and stops short of resolving correlated-feature effects.

As explained in Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI

A partial dependence plot displays a model's average predicted response as one or more input features vary while marginalizing over other features.

A partial dependence plot (PDP) shows how a trained model’s average prediction changes as one or more input features vary, while marginalizing over other features. It helps reveal modeled linearities, nonlinearities, and possible interactions before transforming numerical data. section 6.1.4

Use PDPs with these practical cautions:

Interpret the plotted feature as having a direct effect only when it is independent of other inputs; correlated features make the result an average that also reflects their effects. section 6.1.4
Compare average PDPs with ICE plots when individual prediction paths matter, and inspect two-feature PDPs when interactions are suspected. section 6.1.4

Airbnb NYC model inspection

Fit an XGBoost model in a preprocessing pipeline, then call scikit-learn’s PartialDependenceDisplay with kind='average' for features such as minimum_nights, number_of_reviews, calculated_host_listings_count, and availability_365. Plot a classification threshold to help interpret predicted outcomes.

How Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI covers it

Cover of Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI

liveBook manning.com

Read Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI on liveBook Get the book at manning.com

Also explained in

liveBook manning.com

Ensemble Methods for Machine LearningA partial dependence function is the average model prediction obtained by fixing selected feature values and integrating or averaging over the remaining features.

Compare how 2 books explain partial dependence plot →

Related concepts

marginalizationused by · Marginalization is the operation of summing or integrating over variables or alternatives that are not retained to obtain a distribution or quantity for the remaining variables.
elasticity plotincludes · An elasticity plot shows how a model's predicted outcome changes as a price-related input is varied for a product or SKU.
individual conditional expectation plotincludes · An individual conditional expectation plot shows how a model’s prediction changes for an individual observation as a feature varies.
training datasetuses · A training dataset is a collection of examples or data used by a learning procedure to fit or adapt a model.

This explainer was generated by Manning's knowledge graph from the text of Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI. The section links point to the parts of the book it draws on.