How deeply does Outlier Detection in Python cover predict_proba?

Defines output shape and default scaling, but stops short of treating scores as calibrated probabilities.

Predict_proba

As explained in Outlier Detection in Python

predict_proba is an estimator method that returns normalized class-membership scores for each supplied record.

predict_proba() returns a two-column array of normalized scores for each record: estimated inlier and outlier membership. In PyOD, these values are scaled scores rather than true probabilities, although they can be interpreted similarly. section 6.1

Interpret the method alongside the other common detector operations:

fit() trains the detector; predict() returns binary labels; decision_function() returns numeric outlier scores. section 6.1
By default, predict_proba() applies min-max scaling to decision_function() results, placing scores between 0.0 and 1.0. section 6.1
Each row contains inlier and outlier scores, which sum to 1.0; alternative scaling methods are available. section 6.1

How Outlier Detection in Python covers it

liveBook manning.com

Read Outlier Detection in Python on liveBook Get the book at manning.com

Related concepts

class probability distributionproduces · A class probability distribution is a set of estimated probabilities assigned to the possible classes for an input, typically nonnegative and summing to one.
confidencemeasures · Confidence is a degree of certainty or assurance associated with a judgment, estimate, prediction, capability, or statistical procedure, often expressed as a score, probability, or coverage level.
min-max normalizationuses · Min-max normalization is a feature-scaling technique that linearly maps numerical values to a specified interval using their minimum and maximum values.
independent variableused by · An independent variable is an input or explanatory attribute whose values are used to model, explain, or predict variation in an outcome.

This explainer was generated by Manning's knowledge graph from the text of Outlier Detection in Python. The section links point to the parts of the book it draws on.