test_projected_gradient_descent.py

# MIT License
#
# Copyright (C) IBM Corporation 2018
#
# Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated
# documentation files (the "Software"), to deal in the Software without restriction, including without limitation the
# rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit
# persons to whom the Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all copies or substantial portions of the
# Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
# WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from __future__ import absolute_import, division, print_function, unicode_literals

import logging
import unittest

import tensorflow as tf
import keras.backend as k
import numpy as np

from art.attacks import ProjectedGradientDescent
from art.classifiers import KerasClassifier
from art.utils import load_dataset, get_labels_np_array, master_seed, random_targets
from art.utils_test import get_classifier_tf, get_classifier_kr, get_classifier_pt
from art.utils_test import get_iris_classifier_tf, get_iris_classifier_kr, get_iris_classifier_pt

logger = logging.getLogger('testLogger')

BATCH_SIZE = 10
NB_TRAIN = 100
NB_TEST = 11


@unittest.skipIf(tf.__version__[0] == '2', reason='Skip unittests for TensorFlow v2 until Keras supports TensorFlow'
                                                  ' v2 as backend.')
class TestPGD(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        k.set_learning_phase(1)

        # Get MNIST
        (x_train, y_train), (x_test, y_test), _, _ = load_dataset('mnist')
        x_train, y_train, x_test, y_test = x_train[:NB_TRAIN], y_train[:NB_TRAIN], x_test[:NB_TEST], y_test[:NB_TEST]
        cls.mnist = (x_train, y_train), (x_test, y_test)

        # Keras classifier
        cls.classifier_k = get_classifier_kr()

        scores = cls.classifier_k._model.evaluate(x_train, y_train)
        logger.info('[Keras, MNIST] Accuracy on training set: %.2f%%', scores[1] * 100)
        scores = cls.classifier_k._model.evaluate(x_test, y_test)
        logger.info('[Keras, MNIST] Accuracy on test set: %.2f%%', scores[1] * 100)

        # Create basic CNN on MNIST using TensorFlow
        cls.classifier_tf, sess = get_classifier_tf()

        scores = get_labels_np_array(cls.classifier_tf.predict(x_train))
        acc = np.sum(np.argmax(scores, axis=1) == np.argmax(y_train, axis=1)) / y_train.shape[0]
        logger.info('[TF, MNIST] Accuracy on training set: %.2f%%', acc * 100)

        scores = get_labels_np_array(cls.classifier_tf.predict(x_test))
        acc = np.sum(np.argmax(scores, axis=1) == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('[TF, MNIST] Accuracy on test set: %.2f%%', acc * 100)

        # Create basic PyTorch model
        cls.classifier_py = get_classifier_pt()
        x_train, x_test = np.swapaxes(x_train, 1, 3), np.swapaxes(x_test, 1, 3)
        x_train, x_test = x_train.astype(np.float32), x_test.astype(np.float32)

        scores = get_labels_np_array(cls.classifier_py.predict(x_train))
        acc = np.sum(np.argmax(scores, axis=1) == np.argmax(y_train, axis=1)) / y_train.shape[0]
        logger.info('[PyTorch, MNIST] Accuracy on training set: %.2f%%', acc * 100)

        scores = get_labels_np_array(cls.classifier_py.predict(x_test))
        acc = np.sum(np.argmax(scores, axis=1) == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('[PyTorch, MNIST] Accuracy on test set: %.2f%%', acc * 100)

    def setUp(self):
        # Set master seed
        master_seed(1234)

    def test_mnist(self):
        # Define all backends to test
        backends = {'keras': self.classifier_k,
                    'tf': self.classifier_tf,
                    'pytorch': self.classifier_py}

        for _, classifier in backends.items():
            if _ == 'pytorch':
                self._swap_axes()
                (x_train, y_train), (x_test, y_test) = self.mnist
                self.mnist = (x_train.astype(np.float32), y_train), (x_test.astype(np.float32), y_test)

            self._test_backend_mnist(classifier)

            if _ == 'pytorch':
                self._swap_axes()
                (x_train, y_train), (x_test, y_test) = self.mnist
                self.mnist = (x_train.astype(np.float64), y_train), (x_test.astype(np.float64), y_test)

    def _swap_axes(self):
        (x_train, y_train), (x_test, y_test) = self.mnist
        x_train = np.swapaxes(x_train, 1, 3)
        x_test = np.swapaxes(x_test, 1, 3)
        self.mnist = (x_train, y_train), (x_test, y_test)

    def _test_backend_mnist(self, classifier):
        # Get MNIST
        (x_train, y_train), (x_test, y_test) = self.mnist

        # Test PGD with np.inf norm
        attack = ProjectedGradientDescent(classifier, eps=1, eps_step=0.1)
        x_train_adv = attack.generate(x_train)
        x_test_adv = attack.generate(x_test)

        self.assertFalse((x_train == x_train_adv).all())
        self.assertFalse((x_test == x_test_adv).all())

        train_y_pred = get_labels_np_array(classifier.predict(x_train_adv))
        test_y_pred = get_labels_np_array(classifier.predict(x_test_adv))

        self.assertFalse((y_train == train_y_pred).all())
        self.assertFalse((y_test == test_y_pred).all())

        acc = np.sum(np.argmax(train_y_pred, axis=1) == np.argmax(y_train, axis=1)) / y_train.shape[0]
        logger.info('Accuracy on adversarial train examples: %.2f%%', acc * 100)

        acc = np.sum(np.argmax(test_y_pred, axis=1) == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('Accuracy on adversarial test examples: %.2f%%', acc * 100)

        # Test PGD with 3 random initialisations
        attack = ProjectedGradientDescent(classifier, num_random_init=3)
        x_train_adv = attack.generate(x_train)
        x_test_adv = attack.generate(x_test)

        self.assertFalse((x_train == x_train_adv).all())
        self.assertFalse((x_test == x_test_adv).all())

        train_y_pred = get_labels_np_array(classifier.predict(x_train_adv))
        test_y_pred = get_labels_np_array(classifier.predict(x_test_adv))

        self.assertFalse((y_train == train_y_pred).all())
        self.assertFalse((y_test == test_y_pred).all())

        acc = np.sum(np.argmax(train_y_pred, axis=1) == np.argmax(y_train, axis=1)) / y_train.shape[0]
        logger.info('Accuracy on adversarial train examples with 3 random initialisations: %.2f%%', acc * 100)

        acc = np.sum(np.argmax(test_y_pred, axis=1) == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('Accuracy on adversarial test examples with 3 random initialisations: %.2f%%', acc * 100)

    def test_classifier_type_check_fail_classifier(self):
        # Use a useless test classifier to test basic classifier properties
        class ClassifierNoAPI:
            pass

        classifier = ClassifierNoAPI
        with self.assertRaises(TypeError) as context:
            _ = ProjectedGradientDescent(classifier=classifier)

        self.assertIn('For `ProjectedGradientDescent` classifier must be an instance of '
                      '`art.classifiers.classifier.Classifier`, the provided classifier is instance of '
                      '(<class \'object\'>,).', str(context.exception))

    def test_classifier_type_check_fail_gradients(self):
        # Use a test classifier not providing gradients required by white-box attack
        from art.classifiers.scikitlearn import ScikitlearnDecisionTreeClassifier
        from sklearn.tree import DecisionTreeClassifier

        classifier = ScikitlearnDecisionTreeClassifier(model=DecisionTreeClassifier())
        with self.assertRaises(TypeError) as context:
            _ = ProjectedGradientDescent(classifier=classifier)

        self.assertIn('For `ProjectedGradientDescent` classifier must be an instance of '
                      '`art.classifiers.classifier.ClassifierGradients`, the provided classifier is instance of '
                      '(<class \'art.classifiers.scikitlearn.ScikitlearnClassifier\'>,).', str(context.exception))


class TestPGDVectors(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        # Get Iris
        (x_train, y_train), (x_test, y_test), _, _ = load_dataset('iris')
        cls.iris = (x_train, y_train), (x_test, y_test)

    def setUp(self):
        master_seed(1234)

    @unittest.skipIf(tf.__version__[0] == '2', reason='Skip unittests for TensorFlow v2 until Keras supports TensorFlow'
                                                      ' v2 as backend.')
    def test_iris_k_clipped(self):
        (_, _), (x_test, y_test) = self.iris
        classifier, _ = get_iris_classifier_kr()

        # Test untargeted attack
        attack = ProjectedGradientDescent(classifier, eps=1, eps_step=0.1)
        x_test_adv = attack.generate(x_test)
        self.assertFalse((x_test == x_test_adv).all())
        self.assertTrue((x_test_adv <= 1).all())
        self.assertTrue((x_test_adv >= 0).all())

        preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
        self.assertFalse((np.argmax(y_test, axis=1) == preds_adv).all())
        acc = np.sum(preds_adv == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('Accuracy on Iris with PGD adversarial examples: %.2f%%', (acc * 100))

        # Test targeted attack
        targets = random_targets(y_test, nb_classes=3)
        attack = ProjectedGradientDescent(classifier, targeted=True, eps=1, eps_step=0.1)
        x_test_adv = attack.generate(x_test, **{'y': targets})
        self.assertFalse((x_test == x_test_adv).all())
        self.assertTrue((x_test_adv <= 1).all())
        self.assertTrue((x_test_adv >= 0).all())

        preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
        self.assertTrue((np.argmax(targets, axis=1) == preds_adv).any())
        acc = np.sum(preds_adv == np.argmax(targets, axis=1)) / y_test.shape[0]
        logger.info('Success rate of targeted PGD on Iris: %.2f%%', (acc * 100))

    @unittest.skipIf(tf.__version__[0] == '2', reason='Skip unittests for TensorFlow v2 until Keras supports TensorFlow'
                                                      ' v2 as backend.')
    def test_iris_k_unbounded(self):
        (_, _), (x_test, y_test) = self.iris
        classifier, _ = get_iris_classifier_kr()

        # Recreate a classifier without clip values
        classifier = KerasClassifier(model=classifier._model, use_logits=False, channel_index=1)
        attack = ProjectedGradientDescent(classifier, eps=1, eps_step=0.2)
        x_test_adv = attack.generate(x_test)
        self.assertFalse((x_test == x_test_adv).all())
        self.assertTrue((x_test_adv > 1).any())
        self.assertTrue((x_test_adv < 0).any())

        preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
        self.assertFalse((np.argmax(y_test, axis=1) == preds_adv).all())
        acc = np.sum(preds_adv == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('Accuracy on Iris with PGD adversarial examples: %.2f%%', (acc * 100))

    def test_iris_tf(self):
        (_, _), (x_test, y_test) = self.iris
        classifier, _ = get_iris_classifier_tf()

        # Test untargeted attack
        attack = ProjectedGradientDescent(classifier, eps=1, eps_step=0.1)
        x_test_adv = attack.generate(x_test)
        self.assertFalse((x_test == x_test_adv).all())
        self.assertTrue((x_test_adv <= 1).all())
        self.assertTrue((x_test_adv >= 0).all())

        preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
        self.assertFalse((np.argmax(y_test, axis=1) == preds_adv).all())
        acc = np.sum(preds_adv == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('Accuracy on Iris with PGD adversarial examples: %.2f%%', (acc * 100))

        # Test targeted attack
        targets = random_targets(y_test, nb_classes=3)
        attack = ProjectedGradientDescent(classifier, targeted=True, eps=1, eps_step=0.1)
        x_test_adv = attack.generate(x_test, **{'y': targets})
        self.assertFalse((x_test == x_test_adv).all())
        self.assertTrue((x_test_adv <= 1).all())
        self.assertTrue((x_test_adv >= 0).all())

        preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
        self.assertTrue((np.argmax(targets, axis=1) == preds_adv).any())
        acc = np.sum(preds_adv == np.argmax(targets, axis=1)) / y_test.shape[0]
        logger.info('Success rate of targeted PGD on Iris: %.2f%%', (acc * 100))

    def test_iris_pt(self):
        (_, _), (x_test, y_test) = self.iris
        classifier = get_iris_classifier_pt()

        # Test untargeted attack
        attack = ProjectedGradientDescent(classifier, eps=1, eps_step=0.1)
        x_test_adv = attack.generate(x_test)
        self.assertFalse((x_test == x_test_adv).all())
        self.assertTrue((x_test_adv <= 1).all())
        self.assertTrue((x_test_adv >= 0).all())

        preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
        self.assertFalse((np.argmax(y_test, axis=1) == preds_adv).all())
        acc = np.sum(preds_adv == np.argmax(y_test, axis=1)) / y_test.shape[0]
        logger.info('Accuracy on Iris with PGD adversarial examples: %.2f%%', (acc * 100))

        # Test targeted attack
        targets = random_targets(y_test, nb_classes=3)
        attack = ProjectedGradientDescent(classifier, targeted=True, eps=1, eps_step=0.1)
        x_test_adv = attack.generate(x_test, **{'y': targets})
        self.assertFalse((x_test == x_test_adv).all())
        self.assertTrue((x_test_adv <= 1).all())
        self.assertTrue((x_test_adv >= 0).all())

        preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
        self.assertTrue((np.argmax(targets, axis=1) == preds_adv).any())
        acc = np.sum(preds_adv == np.argmax(targets, axis=1)) / y_test.shape[0]
        logger.info('Success rate of targeted PGD on Iris: %.2f%%', (acc * 100))

    def test_scikitlearn(self):
        from sklearn.linear_model import LogisticRegression
        from sklearn.svm import SVC, LinearSVC

        from art.classifiers.scikitlearn import ScikitlearnLogisticRegression, ScikitlearnSVC

        scikitlearn_test_cases = {LogisticRegression: ScikitlearnLogisticRegression,
                                  SVC: ScikitlearnSVC,
                                  LinearSVC: ScikitlearnSVC}

        (_, _), (x_test, y_test) = self.iris

        for (model_class, classifier_class) in scikitlearn_test_cases.items():
            model = model_class()
            classifier = classifier_class(model=model, clip_values=(0, 1))
            classifier.fit(x=x_test, y=y_test)

            # Test untargeted attack
            attack = ProjectedGradientDescent(classifier, eps=1, eps_step=0.1)
            x_test_adv = attack.generate(x_test)
            self.assertFalse((x_test == x_test_adv).all())
            self.assertTrue((x_test_adv <= 1).all())
            self.assertTrue((x_test_adv >= 0).all())

            preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
            self.assertFalse((np.argmax(y_test, axis=1) == preds_adv).all())
            acc = np.sum(preds_adv == np.argmax(y_test, axis=1)) / y_test.shape[0]
            logger.info('Accuracy of ' + classifier.__class__.__name__ + ' on Iris with PGD adversarial examples: '
                                                                         '%.2f%%', (acc * 100))

            # Test targeted attack
            targets = random_targets(y_test, nb_classes=3)
            attack = ProjectedGradientDescent(classifier, targeted=True, eps=1, eps_step=0.1)
            x_test_adv = attack.generate(x_test, **{'y': targets})
            self.assertFalse((x_test == x_test_adv).all())
            self.assertTrue((x_test_adv <= 1).all())
            self.assertTrue((x_test_adv >= 0).all())

            preds_adv = np.argmax(classifier.predict(x_test_adv), axis=1)
            self.assertTrue((np.argmax(targets, axis=1) == preds_adv).any())
            acc = np.sum(preds_adv == np.argmax(targets, axis=1)) / y_test.shape[0]
            logger.info('Success rate of ' + classifier.__class__.__name__ + ' on targeted PGD on Iris: %.2f%%',
                        (acc * 100))


if __name__ == '__main__':
    unittest.main()