Sklearn Naive Bayes icon

Sklearn Naive Bayes

Naive Bayes classifiers using scikit-learn

Actions2

Train

AI-generated

Summary

Train a Naive Bayes classifier using scikit-learn.

Inputs

  • variant — Type of Naive Bayes classifier: Gaussian NB (continuous features), Multinomial NB (discrete counts), or Bernoulli NB (binary/binary features). Default is Gaussian NB.
  • featureColumns (required) — Comma‑separated list of column names to use as features for training.
  • targetColumn (required) — Name of the column containing the target variable to predict.
  • alpha — Additive (Laplace) smoothing parameter. Only applies to Multinomial NB and Bernoulli NB. Default is 1.
  • pythonPath — Path to the Python executable that has scikit‑learn installed. Default is 'python3'.

Output shape

object

Each execution returns a single item whose JSON contains the following fields: model_pickle (base64‑encoded pickled model), variant (selected variant), score (training accuracy), classes (list of class labels), class_prior (list of class priors or null), feature_columns (list of feature column names used), training_samples (number of input rows).

Examples

Example 1: Train a Gaussian NB on numeric features

variant: gaussian; featureColumns: 'age,income'; targetColumn: 'purchased'

Discussion