DaalNaiveBayesModel train

---

train(self, frame, label_column, observation_columns, num_classes=2)

[BETA] Build a multinomial naive bayes model.

Parameters:

frame : Frame

A frame to train the model on.

label_column : unicode

Column containing the label for each observation.

observation_columns : list

Column(s) containing the observations.

num_classes : int32 (default=2)

Number of classes

Returns:

: dict

dictionary A dictionary with trained multinomial naive bayes model with the following keys: ‘class_log_prior’: smoothed empirical log probability for each class ‘feature_log_prob’: empirical log probability of features given a class, P(x_i|y)

Train a DaalNaiveBayesModel using the observation column, label column of the train frame and an optional lambda value.

Examples

See here for examples.