LibsvmModel test

---

test(self, frame, label_column, observation_columns=None)

[ALPHA] Predict test frame labels and return metrics.

Parameters:

frame : <bound method AtkEntityType.__name__ of <trustedanalytics.rest.jsonschema.AtkEntityType object at 0x7f9e686f3fd0>>

A frame whose labels are to be predicted.

label_column : unicode

Column containing the actual label for each observation.

observation_columns : list (default=None)

Column(s) containing the observations whose labels are to be predicted and tested. Default is to test over the columns the LIBSVM model was trained on.

Returns:

: dict

Object with binary classification metrics. The data returned is composed of multiple components:

double : accuracy

The degree of correctness of the test frame labels.

table : confusion_matrix

A specific table layout that allows visualization of the performance of the

test. | double : f_measure | A measure of a test’s accuracy. It considers both the precision and the recall of the test to compute the score. | double : precision | The degree to which the correctness of the label is expressed. | double : recall | The fraction of relevant instances that are retrieved.

Predict the labels for a test frame and run classification metrics on predicted and target labels.

Examples

>>> my_model = ta.LibsvmModel(name='mySVM')
>>> my_model.train(train_frame, 'name_of_label_column',['List_of_observation_column/s'])
>>> metrics = my_model.test(test_frame, 'name_of_label_column',['List_of_observation_column/s'])

>>> metrics.f_measure
0.66666666666666663

>>> metrics.recall
0.5

>>> metrics.accuracy
0.75

>>> metrics.precision
1.0

>>> metrics.confusion_matrix

              Predicted
            _pos_ _neg__
Actual  pos |  1     1
        neg |  0     2