def test_samples_fashionMNIST():
images, labels = samples(dataset='fashionMNIST', batchsize=5)
assert 0 <= labels[0] < 10
assert images.shape[0] == 5
assert isinstance(labels[0], np.integer)
assert images.shape == (5, 28, 28)
assert images.dtype == np.float32
def test_samples_imagenet():
images, labels = samples(dataset='imagenet', batchsize=5)
assert 0 <= labels[0] < 1000
assert images.shape[0] == 5
assert isinstance(labels[0], np.integer)
assert images.shape == (5, 224, 224, 3)
assert images.dtype == np.float32
def test_samples_cifar100():
images, labels = samples(dataset='cifar100', batchsize=5)
assert 0 <= labels[0] < 100
assert images.shape[0] == 5
assert isinstance(labels[0], np.integer)
assert images.shape == (5, 32, 32, 3)
assert images.dtype == np.float32
def test_samples_imagenet_channels_first():
images, labels = samples(dataset='imagenet',
batchsize=5,
data_format='channels_first')
assert 0 <= labels[0] < 1000
assert images.shape[0] == 5
assert isinstance(labels[0], np.integer)
assert images.shape == (5, 3, 224, 224)
assert images.dtype == np.float32
def test_samples_mnist():
images, labels = samples(dataset="mnist", batchsize=5)
assert 0 <= labels[0] < 10
assert images.shape[0] == 5
assert isinstance(labels[0], np.integer)
assert images.shape == (5, 28, 28)
assert images.dtype == np.float32
assert 0 <= images.min() <= 5
assert 250 < images.max() <= 255
def test_samples_imagenet_bounds():
images, labels = samples(dataset="imagenet", batchsize=5, bounds=(0, 1))
assert 0 <= labels[0] < 1000
assert images.shape[0] == 5
assert isinstance(labels[0], np.integer)
assert images.shape == (5, 224, 224, 3)
assert images.dtype == np.float32
assert np.all(images >= 0)
assert np.all(images <= 1)
def test_samples_mnist_bounds():
images, labels = samples(dataset="mnist", batchsize=5, bounds=(-1, 1))
assert 0 <= labels[0] < 10
assert images.shape[0] == 5
assert isinstance(labels[0], np.integer)
assert images.shape == (5, 28, 28)
assert images.dtype == np.float32
assert np.all(images >= -1)
assert np.all(images <= 1)
assert images.min() < 0
assert images.max() > 0
def create() -> PyTorchModel:
model = nn.Sequential(
nn.Conv2d(1, 32, 3),
nn.ReLU(),
nn.Conv2d(32, 64, 3),
nn.ReLU(),
nn.MaxPool2d(2),
nn.Dropout2d(0.25),
nn.Flatten(), # type: ignore
nn.Linear(9216, 128),
nn.ReLU(),
nn.Dropout2d(0.5),
nn.Linear(128, 10),
)
path = os.path.join(os.path.dirname(os.path.realpath(__file__)),
"mnist_cnn.pth")
model.load_state_dict(torch.load(path)) # type: ignore
model.eval()
preprocessing = dict(mean=0.1307, std=0.3081)
fmodel = PyTorchModel(model, bounds=(0, 1), preprocessing=preprocessing)
return fmodel
if __name__ == "__main__":
# test the model
fmodel = create()
images, labels = samples(fmodel, dataset="mnist", batchsize=20)
print(accuracy(fmodel, images, labels))
print("Load Config")
if args.config_path is not None:
if not os.path.exists(args.config_path):
raise ValueError("{} doesn't exist.".format(args.config_path))
config = json.load(open(args.config_path, "r"))
else:
config = {"init": {}, "run": {"epsilons": None}}
###############################
print("Get understandable ImageNet labels")
imagenet_labels = get_imagenet_labels()
###############################
print("Load Data")
images, labels = samples(fmodel,
dataset="imagenet",
batchsize=args.n_images)
print("{} images loaded with the following labels: {}".format(
len(images), labels))
###############################
print("Attack !")
time_start = time.time()
f_attack = SurFree(**config["init"])
elements, advs, success = f_attack(fmodel, images, labels, **config["run"])
print("{:.2f} s to run".format(time.time() - time_start))
###############################
print("Results")