model Package

model Package

auc_data Module

An object containing information about a binomial classifier.

class h2o.model.auc_data.AUCData(raw_auc)

Bases: object

class h2o.model.auc_data.ThresholdCriterion

Bases: object

An Enum for the Threshold Criteria

ACCURACY = 'maximum Accuracy'

F0POINT5 = 'maximum F0point5'

MAXF1 = 'maximum F1'

MAXF2 = 'maximum F2'

MCC = 'maximum absolute MCC'

MINMAXPERCLASSERR = 'minimizing max per class Error'

PRECISION = 'maximum Precision'

RECALL = 'maximum Recall'

SPECIFICITY = 'maximum Specificity'

crits()

binomial Module

Binomial Models should be comparable.

class h2o.model.binomial.H2OBinomialModel(dest_key, model_json)

Bases: h2o.model.model_base.ModelBase

Class for Binomial models.

class h2o.model.binomial.H2OBinomialModelMetrics(metric_json)

Bases: object

This class is essentially an API for the AUCData object. This class contains methods for inspecting the AUC for different criteria. To input the different criteria, use the static variable criteria

F0point5(thresholds=None)

F1(thresholds=None)

F2(thresholds=None)

accuracy(thresholds=None)

auc()

confusion_matrices(thresholds=None)

error(thresholds=None)

giniCoef()

max_per_class_error(thresholds=None)

mcc(thresholds=None)

metric(metric='accuracy', thresholds=None)

mse()

precision(thresholds=None)

recall(thresholds=None)

show()

specificity(thresholds=None)

theCriteria = <h2o.model.auc_data.ThresholdCriterion object at 0x4d10690>

clustering Module

Clustering Models should be comparable.

class h2o.model.clustering.H2OClusteringModel(dest_key, model_json)

Bases: h2o.model.model_base.ModelBase

avg_between_ss()

avg_ss()

avg_within_ss()

centers()

size()

summary()

This method prints out various relevant pieces of information for a clustering model.

within_mse()

class h2o.model.clustering.H2OClusteringModelMetrics(metric_json)

Bases: object

confusion_matrix Module

A confusion matrix from H2O.

class h2o.model.confusion_matrix.ConfusionMatrix(cm, domains=None)

Bases: object

ROUND = 4

static read_cms(cms=None, domains=None)

show()

model_base Module

This module implements the base model class. All model things inherit from this class.

class h2o.model.model_base.ModelBase(dest_key, model_json, metrics_class)

Bases: object

model_performance(test_data)

Generate model metrics for this model on test_data. :param test_data: Data set for which model metrics shall be computed against. :return: An object of class H2OModelMetrics.

predict(test_data)

Predict on a dataset. :param test_data: Data to be predicted on. :return: A new H2OFrame filled with predictions.

show()

Print innards of model, without regards to type :return: None

summary()

Print a detailed summary of the model. :return:

multinomial Module

Multinomial Models should be comparable.

class h2o.model.multinomial.H2OMultinomialModel(dest_key, model_json)

Bases: h2o.model.model_base.ModelBase

summary()

This method prints out various relevant pieces of information for a multinomial model. :return:

class h2o.model.multinomial.H2OMultinomialModelMetrics(metric_json)

Bases: object

regression Module

Regression Models should be comparable.

class h2o.model.regression.H2ORegressionModel(dest_key, model_json)

Bases: h2o.model.model_base.ModelBase

Class for Regression models.

class h2o.model.regression.H2ORegressionModelMetrics(metric_json)

Bases: object

This class provides an API for inspecting the metrics returned by a regression model.

It is possible to retrieve the R^2 (1 - mse/variance), mse, and sigma.s

mse()

Returns:The MSE for this regression model.

r2()

Return the R^2 for this regression model.

The R^2 value is defined to be 1 - mse/var, where var is computed as sigma*sigma. :return: The R^2 for this regression model.

show()

sigma()

Returns: