4 examples of 'scikit learn classification report' in Python

405
def runClassReport(clf, A, Cl):
406
    from sklearn.metrics import classification_report
407
    y_pred = clf.predict(A)
408
    print(classification_report(Cl, y_pred, target_names=clf.classes_))
409
    print(' Precision is the probability that, given a classification result for a sample,\n' +
410
          ' the sample actually belongs to that class. Recall (Accuracy) is the probability that a \n' +
411
          ' sample will be correctly classified for a given class. f1-score combines both \n' +
412
          ' accuracy and precision to give a single measure of relevancy of the classifier results.\n')

60
def _check_classification_report(n_classes=2):
61
    classifiers = ClassifiersFactory()
62
    classifiers.add_classifier('gb', GradientBoostingClassifier(n_estimators=10))
63
    classifiers.add_classifier('rf', RandomForestClassifier())
64
    classifiers.add_classifier('ada', AdaBoostClassifier(n_estimators=10))
65
66
    X, y = generate_classification_sample(1000, 5, n_classes=n_classes)
67
    classifiers.fit(X, y)
68
69
    X, y = generate_classification_sample(1000, 5, n_classes=n_classes)
70
    test_lds = LabeledDataStorage(X, y, sample_weight=None)
71
    report = classifiers.test_on_lds(test_lds)
72
73
    val = numpy.mean(X['column0'])
74
    labels_dict = None
75
    if n_classes > 2:
76
        labels_dict = {i: str(i) for i in range(n_classes)}
77
    _classification_mask_report(report, "column0 > %f" % val, X, labels_dict)
78
    _classification_mask_report(report, lambda x: numpy.array(x['column0']) < val, X, labels_dict)
79
    _classification_mask_report(report, None, X, labels_dict)
80
    check_classification_learning_curve_masks(report, n_classes=n_classes)

17
def report(test_Y, pred_Y):
18
    print "accuracy_score:"
19
    print metrics.accuracy_score(test_Y, pred_Y)
20
    print "f1_score:"
21
    print metrics.f1_score(test_Y, pred_Y)
22
    print "recall_score:"
23
    print metrics.recall_score(test_Y, pred_Y)
24
    print "precision_score:"
25
    print metrics.precision_score(test_Y, pred_Y)
26
    print "confusion_matrix:"
27
    print metrics.confusion_matrix(test_Y, pred_Y)
28
    print "AUC:"
29
    print metrics.roc_auc_score(test_Y, pred_Y)
30
31
    f_pos, t_pos, thresh = metrics.roc_curve(test_Y, pred_Y)
32
    auc_area = metrics.auc(f_pos, t_pos)
33
    plt.plot(f_pos, t_pos, 'darkorange', lw=2, label='AUC = %.2f' % auc_area)
34
    plt.legend(loc='lower right')
35
    plt.plot([0, 1], [0, 1], color='navy', linestyle='--')
36
    plt.title('ROC')
37
    plt.ylabel('True Pos Rate')
38
    plt.xlabel('False Pos Rate')
39
    plt.show()

83
def test_classification_report():
84
    _check_classification_report(n_classes=2)