Glossary

As an interdisciplinary field, machine learning takes terminology from statistics, computer science and the various fields in which it is commonly employed. Sometimes this results in multiple terms being used to refer to the same concept, or very similar concepts.

The following is a table of terms used in the scikit-learn API and documentation, with various (near-)synonyms used in machine learning and related fields.

scikit-learn term	alternatives
coefficient	weight, parameter
estimator	learner
feature	attribute, independent variable, input variable, predictor, regressor
intercept	bias, offset
logistic regression	logit regression, maximum entropy classifier
one-vs.-rest	binary relevance, one-vs.-all
sample	instance, point, event