y_min=-100,
y_max=3000,
steps=750,
pair_wise=[3, 1],
train_or_test=0,
smoothness=0.1,
contour=False)
""" Instantiate an object of Evaluation class to calculate various model metrics. """
eval = Evaluation(bayes_case=bayes_classifier,
data_prep=data_preprocess,
test_size=0.30)
class_id = 1 # The class_id for the required class
# Returns confusion matrix for a given Bayesian Classifier Case
cm = eval.confusion_matrix()
# Returns the accuracy of classification for a given Bayesian Classifier Case
acc = eval.accuracy()
# Returns the precision for a given class for a given Bayesian Classifier Case
prec = eval.precision(class_id)
# Returns the recall for a given class for a given Bayesian Classifier Case
rec = eval.recall(class_id)
# Returns the F-score for a given class for a given Bayesian Classifier Case
f_score = eval.f_score(class_id)
# Returns the mean precision of classification for a given Bayesian Classifier Case
mean_prec = eval.mean_precision()
cant_classified, cant_not_classified = evaluation.evaluate_tree(
tree, data_set_test)
print('De las {} instancias tomadas para evaluar:'.format(cant_classified +
cant_not_classified))
print('\t -> {} instancias clasificaron correctamente'.format(cant_classified))
print('\t -> {} instancias clasificaron incorrectamente'.format(
cant_not_classified))
print('\n')
print('Árboles de clases generados: ')
classes_trees = []
for label in data_set.target_values():
data_set_class = data_set.data_set_class(label)
attributes_aux = attributes.copy()
tree = id3.generate_class_tree(data_set_class, label, attributes,
attributes_aux)
classes_trees.append((label, tree))
print('Clase: ' + str(label))
tree.print(0)
print('\n')
print('Matriz de confusión: ')
confusion_matrix = evaluation.confusion_matrix(data_set_test, classes_trees)
for i in confusion_matrix:
print(str(i))
