repo
stringclasses
85 values
path
stringlengths
8
121
func_name
stringlengths
1
82
original_string
stringlengths
112
65.5k
language
stringclasses
1 value
code
stringlengths
112
65.5k
code_tokens
listlengths
20
4.09k
docstring
stringlengths
3
46.3k
docstring_tokens
listlengths
1
564
sha
stringclasses
85 values
url
stringlengths
93
218
partition
stringclasses
1 value
tensorflow/cleverhans
cleverhans/future/tf2/attacks/fast_gradient_method.py
compute_gradient
def compute_gradient(model_fn, x, y, targeted): """ Computes the gradient of the loss with respect to the input tensor. :param model_fn: a callable that takes an input tensor and returns the model logits. :param x: input tensor :param y: Tensor with true labels. If targeted is true, then provide the target la...
python
def compute_gradient(model_fn, x, y, targeted): """ Computes the gradient of the loss with respect to the input tensor. :param model_fn: a callable that takes an input tensor and returns the model logits. :param x: input tensor :param y: Tensor with true labels. If targeted is true, then provide the target la...
[ "def", "compute_gradient", "(", "model_fn", ",", "x", ",", "y", ",", "targeted", ")", ":", "loss_fn", "=", "tf", ".", "nn", ".", "sparse_softmax_cross_entropy_with_logits", "with", "tf", ".", "GradientTape", "(", ")", "as", "g", ":", "g", ".", "watch", "...
Computes the gradient of the loss with respect to the input tensor. :param model_fn: a callable that takes an input tensor and returns the model logits. :param x: input tensor :param y: Tensor with true labels. If targeted is true, then provide the target label. :param targeted: bool. Is the attack targeted or...
[ "Computes", "the", "gradient", "of", "the", "loss", "with", "respect", "to", "the", "input", "tensor", ".", ":", "param", "model_fn", ":", "a", "callable", "that", "takes", "an", "input", "tensor", "and", "returns", "the", "model", "logits", ".", ":", "p...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/future/tf2/attacks/fast_gradient_method.py#L66-L87
train
tensorflow/cleverhans
cleverhans/future/tf2/attacks/fast_gradient_method.py
optimize_linear
def optimize_linear(grad, eps, ord=np.inf): """ Solves for the optimal input to a linear function under a norm constraint. Optimal_perturbation = argmax_{eta, ||eta||_{ord} < eps} dot(eta, grad) :param grad: tf tensor containing a batch of gradients :param eps: float scalar specifying size of constraint reg...
python
def optimize_linear(grad, eps, ord=np.inf): """ Solves for the optimal input to a linear function under a norm constraint. Optimal_perturbation = argmax_{eta, ||eta||_{ord} < eps} dot(eta, grad) :param grad: tf tensor containing a batch of gradients :param eps: float scalar specifying size of constraint reg...
[ "def", "optimize_linear", "(", "grad", ",", "eps", ",", "ord", "=", "np", ".", "inf", ")", ":", "# Convert the iterator returned by `range` into a list.", "axis", "=", "list", "(", "range", "(", "1", ",", "len", "(", "grad", ".", "get_shape", "(", ")", ")"...
Solves for the optimal input to a linear function under a norm constraint. Optimal_perturbation = argmax_{eta, ||eta||_{ord} < eps} dot(eta, grad) :param grad: tf tensor containing a batch of gradients :param eps: float scalar specifying size of constraint region :param ord: int specifying order of norm :re...
[ "Solves", "for", "the", "optimal", "input", "to", "a", "linear", "function", "under", "a", "norm", "constraint", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/future/tf2/attacks/fast_gradient_method.py#L90-L128
train
tensorflow/cleverhans
cleverhans/plot/save_pdf.py
save_pdf
def save_pdf(path): """ Saves a pdf of the current matplotlib figure. :param path: str, filepath to save to """ pp = PdfPages(path) pp.savefig(pyplot.gcf()) pp.close()
python
def save_pdf(path): """ Saves a pdf of the current matplotlib figure. :param path: str, filepath to save to """ pp = PdfPages(path) pp.savefig(pyplot.gcf()) pp.close()
[ "def", "save_pdf", "(", "path", ")", ":", "pp", "=", "PdfPages", "(", "path", ")", "pp", ".", "savefig", "(", "pyplot", ".", "gcf", "(", ")", ")", "pp", ".", "close", "(", ")" ]
Saves a pdf of the current matplotlib figure. :param path: str, filepath to save to
[ "Saves", "a", "pdf", "of", "the", "current", "matplotlib", "figure", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/plot/save_pdf.py#L8-L17
train
tensorflow/cleverhans
cleverhans/future/tf2/utils_tf.py
clip_eta
def clip_eta(eta, ord, eps): """ Helper function to clip the perturbation to epsilon norm ball. :param eta: A tensor with the current perturbation. :param ord: Order of the norm (mimics Numpy). Possible values: np.inf, 1 or 2. :param eps: Epsilon, bound of the perturbation. """ # Clipping p...
python
def clip_eta(eta, ord, eps): """ Helper function to clip the perturbation to epsilon norm ball. :param eta: A tensor with the current perturbation. :param ord: Order of the norm (mimics Numpy). Possible values: np.inf, 1 or 2. :param eps: Epsilon, bound of the perturbation. """ # Clipping p...
[ "def", "clip_eta", "(", "eta", ",", "ord", ",", "eps", ")", ":", "# Clipping perturbation eta to self.ord norm ball", "if", "ord", "not", "in", "[", "np", ".", "inf", ",", "1", ",", "2", "]", ":", "raise", "ValueError", "(", "'ord must be np.inf, 1, or 2.'", ...
Helper function to clip the perturbation to epsilon norm ball. :param eta: A tensor with the current perturbation. :param ord: Order of the norm (mimics Numpy). Possible values: np.inf, 1 or 2. :param eps: Epsilon, bound of the perturbation.
[ "Helper", "function", "to", "clip", "the", "perturbation", "to", "epsilon", "norm", "ball", ".", ":", "param", "eta", ":", "A", "tensor", "with", "the", "current", "perturbation", ".", ":", "param", "ord", ":", "Order", "of", "the", "norm", "(", "mimics"...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/future/tf2/utils_tf.py#L5-L33
train
tensorflow/cleverhans
cleverhans_tutorials/mnist_blackbox.py
prep_bbox
def prep_bbox(sess, x, y, x_train, y_train, x_test, y_test, nb_epochs, batch_size, learning_rate, rng, nb_classes=10, img_rows=28, img_cols=28, nchannels=1): """ Define and train a model that simulates the "remote" black-box oracle described in the original paper. :param sess: the TF...
python
def prep_bbox(sess, x, y, x_train, y_train, x_test, y_test, nb_epochs, batch_size, learning_rate, rng, nb_classes=10, img_rows=28, img_cols=28, nchannels=1): """ Define and train a model that simulates the "remote" black-box oracle described in the original paper. :param sess: the TF...
[ "def", "prep_bbox", "(", "sess", ",", "x", ",", "y", ",", "x_train", ",", "y_train", ",", "x_test", ",", "y_test", ",", "nb_epochs", ",", "batch_size", ",", "learning_rate", ",", "rng", ",", "nb_classes", "=", "10", ",", "img_rows", "=", "28", ",", "...
Define and train a model that simulates the "remote" black-box oracle described in the original paper. :param sess: the TF session :param x: the input placeholder for MNIST :param y: the ouput placeholder for MNIST :param x_train: the training data for the oracle :param y_train: the training labels for the ...
[ "Define", "and", "train", "a", "model", "that", "simulates", "the", "remote", "black", "-", "box", "oracle", "described", "in", "the", "original", "paper", ".", ":", "param", "sess", ":", "the", "TF", "session", ":", "param", "x", ":", "the", "input", ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans_tutorials/mnist_blackbox.py#L59-L101
train
tensorflow/cleverhans
cleverhans_tutorials/mnist_blackbox.py
train_sub
def train_sub(sess, x, y, bbox_preds, x_sub, y_sub, nb_classes, nb_epochs_s, batch_size, learning_rate, data_aug, lmbda, aug_batch_size, rng, img_rows=28, img_cols=28, nchannels=1): """ This function creates the substitute by alternatively augmenting the training data and...
python
def train_sub(sess, x, y, bbox_preds, x_sub, y_sub, nb_classes, nb_epochs_s, batch_size, learning_rate, data_aug, lmbda, aug_batch_size, rng, img_rows=28, img_cols=28, nchannels=1): """ This function creates the substitute by alternatively augmenting the training data and...
[ "def", "train_sub", "(", "sess", ",", "x", ",", "y", ",", "bbox_preds", ",", "x_sub", ",", "y_sub", ",", "nb_classes", ",", "nb_epochs_s", ",", "batch_size", ",", "learning_rate", ",", "data_aug", ",", "lmbda", ",", "aug_batch_size", ",", "rng", ",", "im...
This function creates the substitute by alternatively augmenting the training data and training the substitute. :param sess: TF session :param x: input TF placeholder :param y: output TF placeholder :param bbox_preds: output of black-box model predictions :param x_sub: initial substitute training data :pa...
[ "This", "function", "creates", "the", "substitute", "by", "alternatively", "augmenting", "the", "training", "data", "and", "training", "the", "substitute", ".", ":", "param", "sess", ":", "TF", "session", ":", "param", "x", ":", "input", "TF", "placeholder", ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans_tutorials/mnist_blackbox.py#L123-L188
train
tensorflow/cleverhans
cleverhans_tutorials/mnist_blackbox.py
mnist_blackbox
def mnist_blackbox(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_classes=NB_CLASSES, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, nb_epochs=NB_EPOCHS, holdout=HOLDOUT, data_aug=DATA_AUG, nb_epochs_s=NB_EPOCHS_S, lmbda=...
python
def mnist_blackbox(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_classes=NB_CLASSES, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, nb_epochs=NB_EPOCHS, holdout=HOLDOUT, data_aug=DATA_AUG, nb_epochs_s=NB_EPOCHS_S, lmbda=...
[ "def", "mnist_blackbox", "(", "train_start", "=", "0", ",", "train_end", "=", "60000", ",", "test_start", "=", "0", ",", "test_end", "=", "10000", ",", "nb_classes", "=", "NB_CLASSES", ",", "batch_size", "=", "BATCH_SIZE", ",", "learning_rate", "=", "LEARNIN...
MNIST tutorial for the black-box attack from arxiv.org/abs/1602.02697 :param train_start: index of first training set example :param train_end: index of last training set example :param test_start: index of first test set example :param test_end: index of last test set example :return: a dictionary with: ...
[ "MNIST", "tutorial", "for", "the", "black", "-", "box", "attack", "from", "arxiv", ".", "org", "/", "abs", "/", "1602", ".", "02697", ":", "param", "train_start", ":", "index", "of", "first", "training", "set", "example", ":", "param", "train_end", ":", ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans_tutorials/mnist_blackbox.py#L191-L284
train
tensorflow/cleverhans
cleverhans/augmentation.py
random_shift
def random_shift(x, pad=(4, 4), mode='REFLECT'): """Pad a single image and then crop to the original size with a random offset.""" assert mode in 'REFLECT SYMMETRIC CONSTANT'.split() assert x.get_shape().ndims == 3 xp = tf.pad(x, [[pad[0], pad[0]], [pad[1], pad[1]], [0, 0]], mode) return tf.random_crop(xp, ...
python
def random_shift(x, pad=(4, 4), mode='REFLECT'): """Pad a single image and then crop to the original size with a random offset.""" assert mode in 'REFLECT SYMMETRIC CONSTANT'.split() assert x.get_shape().ndims == 3 xp = tf.pad(x, [[pad[0], pad[0]], [pad[1], pad[1]], [0, 0]], mode) return tf.random_crop(xp, ...
[ "def", "random_shift", "(", "x", ",", "pad", "=", "(", "4", ",", "4", ")", ",", "mode", "=", "'REFLECT'", ")", ":", "assert", "mode", "in", "'REFLECT SYMMETRIC CONSTANT'", ".", "split", "(", ")", "assert", "x", ".", "get_shape", "(", ")", ".", "ndims...
Pad a single image and then crop to the original size with a random offset.
[ "Pad", "a", "single", "image", "and", "then", "crop", "to", "the", "original", "size", "with", "a", "random", "offset", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/augmentation.py#L19-L25
train
tensorflow/cleverhans
cleverhans/augmentation.py
batch_augment
def batch_augment(x, func, device='/CPU:0'): """ Apply dataset augmentation to a batch of exmaples. :param x: Tensor representing a batch of examples. :param func: Callable implementing dataset augmentation, operating on a single image. :param device: String specifying which device to use. """ with tf...
python
def batch_augment(x, func, device='/CPU:0'): """ Apply dataset augmentation to a batch of exmaples. :param x: Tensor representing a batch of examples. :param func: Callable implementing dataset augmentation, operating on a single image. :param device: String specifying which device to use. """ with tf...
[ "def", "batch_augment", "(", "x", ",", "func", ",", "device", "=", "'/CPU:0'", ")", ":", "with", "tf", ".", "device", "(", "device", ")", ":", "return", "tf", ".", "map_fn", "(", "func", ",", "x", ")" ]
Apply dataset augmentation to a batch of exmaples. :param x: Tensor representing a batch of examples. :param func: Callable implementing dataset augmentation, operating on a single image. :param device: String specifying which device to use.
[ "Apply", "dataset", "augmentation", "to", "a", "batch", "of", "exmaples", ".", ":", "param", "x", ":", "Tensor", "representing", "a", "batch", "of", "examples", ".", ":", "param", "func", ":", "Callable", "implementing", "dataset", "augmentation", "operating",...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/augmentation.py#L28-L37
train
tensorflow/cleverhans
cleverhans/augmentation.py
random_crop_and_flip
def random_crop_and_flip(x, pad_rows=4, pad_cols=4): """Augment a batch by randomly cropping and horizontally flipping it.""" rows = tf.shape(x)[1] cols = tf.shape(x)[2] channels = x.get_shape()[3] def _rand_crop_img(img): """Randomly crop an individual image""" return tf.random_crop(img, [rows, cols...
python
def random_crop_and_flip(x, pad_rows=4, pad_cols=4): """Augment a batch by randomly cropping and horizontally flipping it.""" rows = tf.shape(x)[1] cols = tf.shape(x)[2] channels = x.get_shape()[3] def _rand_crop_img(img): """Randomly crop an individual image""" return tf.random_crop(img, [rows, cols...
[ "def", "random_crop_and_flip", "(", "x", ",", "pad_rows", "=", "4", ",", "pad_cols", "=", "4", ")", ":", "rows", "=", "tf", ".", "shape", "(", "x", ")", "[", "1", "]", "cols", "=", "tf", ".", "shape", "(", "x", ")", "[", "2", "]", "channels", ...
Augment a batch by randomly cropping and horizontally flipping it.
[ "Augment", "a", "batch", "by", "randomly", "cropping", "and", "horizontally", "flipping", "it", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/augmentation.py#L40-L58
train
tensorflow/cleverhans
cleverhans_tutorials/mnist_tutorial_picklable.py
mnist_tutorial
def mnist_tutorial(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_epochs=NB_EPOCHS, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, clean_train=CLEAN_TRAIN, testing=False, backprop_through_attack=BACKPRO...
python
def mnist_tutorial(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_epochs=NB_EPOCHS, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, clean_train=CLEAN_TRAIN, testing=False, backprop_through_attack=BACKPRO...
[ "def", "mnist_tutorial", "(", "train_start", "=", "0", ",", "train_end", "=", "60000", ",", "test_start", "=", "0", ",", "test_end", "=", "10000", ",", "nb_epochs", "=", "NB_EPOCHS", ",", "batch_size", "=", "BATCH_SIZE", ",", "learning_rate", "=", "LEARNING_...
MNIST cleverhans tutorial :param train_start: index of first training set example :param train_end: index of last training set example :param test_start: index of first test set example :param test_end: index of last test set example :param nb_epochs: number of epochs to train model :param batch_size: size ...
[ "MNIST", "cleverhans", "tutorial", ":", "param", "train_start", ":", "index", "of", "first", "training", "set", "example", ":", "param", "train_end", ":", "index", "of", "last", "training", "set", "example", ":", "param", "test_start", ":", "index", "of", "f...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans_tutorials/mnist_tutorial_picklable.py#L36-L226
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
_project_perturbation
def _project_perturbation(perturbation, epsilon, input_image, clip_min=None, clip_max=None): """Project `perturbation` onto L-infinity ball of radius `epsilon`. Also project into hypercube such that the resulting adversarial example is between clip_min and clip_max, if applicable. """ ...
python
def _project_perturbation(perturbation, epsilon, input_image, clip_min=None, clip_max=None): """Project `perturbation` onto L-infinity ball of radius `epsilon`. Also project into hypercube such that the resulting adversarial example is between clip_min and clip_max, if applicable. """ ...
[ "def", "_project_perturbation", "(", "perturbation", ",", "epsilon", ",", "input_image", ",", "clip_min", "=", "None", ",", "clip_max", "=", "None", ")", ":", "if", "clip_min", "is", "None", "or", "clip_max", "is", "None", ":", "raise", "NotImplementedError", ...
Project `perturbation` onto L-infinity ball of radius `epsilon`. Also project into hypercube such that the resulting adversarial example is between clip_min and clip_max, if applicable.
[ "Project", "perturbation", "onto", "L", "-", "infinity", "ball", "of", "radius", "epsilon", ".", "Also", "project", "into", "hypercube", "such", "that", "the", "resulting", "adversarial", "example", "is", "between", "clip_min", "and", "clip_max", "if", "applicab...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L209-L231
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
margin_logit_loss
def margin_logit_loss(model_logits, label, nb_classes=10, num_classes=None): """Computes difference between logit for `label` and next highest logit. The loss is high when `label` is unlikely (targeted by default). This follows the same interface as `loss_fn` for TensorOptimizer and projected_optimization, i.e...
python
def margin_logit_loss(model_logits, label, nb_classes=10, num_classes=None): """Computes difference between logit for `label` and next highest logit. The loss is high when `label` is unlikely (targeted by default). This follows the same interface as `loss_fn` for TensorOptimizer and projected_optimization, i.e...
[ "def", "margin_logit_loss", "(", "model_logits", ",", "label", ",", "nb_classes", "=", "10", ",", "num_classes", "=", "None", ")", ":", "if", "num_classes", "is", "not", "None", ":", "warnings", ".", "warn", "(", "\"`num_classes` is depreciated. Switch to `nb_clas...
Computes difference between logit for `label` and next highest logit. The loss is high when `label` is unlikely (targeted by default). This follows the same interface as `loss_fn` for TensorOptimizer and projected_optimization, i.e. it returns a batch of loss values.
[ "Computes", "difference", "between", "logit", "for", "label", "and", "next", "highest", "logit", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L444-L473
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
spm
def spm(x, model, y=None, n_samples=None, dx_min=-0.1, dx_max=0.1, n_dxs=5, dy_min=-0.1, dy_max=0.1, n_dys=5, angle_min=-30, angle_max=30, n_angles=31, black_border_size=0): """ TensorFlow implementation of the Spatial Transformation Method. :return: a tensor for the adversarial example """ if...
python
def spm(x, model, y=None, n_samples=None, dx_min=-0.1, dx_max=0.1, n_dxs=5, dy_min=-0.1, dy_max=0.1, n_dys=5, angle_min=-30, angle_max=30, n_angles=31, black_border_size=0): """ TensorFlow implementation of the Spatial Transformation Method. :return: a tensor for the adversarial example """ if...
[ "def", "spm", "(", "x", ",", "model", ",", "y", "=", "None", ",", "n_samples", "=", "None", ",", "dx_min", "=", "-", "0.1", ",", "dx_max", "=", "0.1", ",", "n_dxs", "=", "5", ",", "dy_min", "=", "-", "0.1", ",", "dy_max", "=", "0.1", ",", "n_...
TensorFlow implementation of the Spatial Transformation Method. :return: a tensor for the adversarial example
[ "TensorFlow", "implementation", "of", "the", "Spatial", "Transformation", "Method", ".", ":", "return", ":", "a", "tensor", "for", "the", "adversarial", "example" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L524-L581
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
parallel_apply_transformations
def parallel_apply_transformations(x, transforms, black_border_size=0): """ Apply image transformations in parallel. :param transforms: TODO :param black_border_size: int, size of black border to apply Returns: Transformed images """ transforms = tf.convert_to_tensor(transforms, dtype=tf.float32) x ...
python
def parallel_apply_transformations(x, transforms, black_border_size=0): """ Apply image transformations in parallel. :param transforms: TODO :param black_border_size: int, size of black border to apply Returns: Transformed images """ transforms = tf.convert_to_tensor(transforms, dtype=tf.float32) x ...
[ "def", "parallel_apply_transformations", "(", "x", ",", "transforms", ",", "black_border_size", "=", "0", ")", ":", "transforms", "=", "tf", ".", "convert_to_tensor", "(", "transforms", ",", "dtype", "=", "tf", ".", "float32", ")", "x", "=", "_apply_black_bord...
Apply image transformations in parallel. :param transforms: TODO :param black_border_size: int, size of black border to apply Returns: Transformed images
[ "Apply", "image", "transformations", "in", "parallel", ".", ":", "param", "transforms", ":", "TODO", ":", "param", "black_border_size", ":", "int", "size", "of", "black", "border", "to", "apply", "Returns", ":", "Transformed", "images" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L584-L610
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
projected_optimization
def projected_optimization(loss_fn, input_image, label, epsilon, num_steps, clip_min=None, clip_max=None, optimizer=TensorAdam(), ...
python
def projected_optimization(loss_fn, input_image, label, epsilon, num_steps, clip_min=None, clip_max=None, optimizer=TensorAdam(), ...
[ "def", "projected_optimization", "(", "loss_fn", ",", "input_image", ",", "label", ",", "epsilon", ",", "num_steps", ",", "clip_min", "=", "None", ",", "clip_max", "=", "None", ",", "optimizer", "=", "TensorAdam", "(", ")", ",", "project_perturbation", "=", ...
Generic projected optimization, generalized to work with approximate gradients. Used for e.g. the SPSA attack. Args: :param loss_fn: A callable which takes `input_image` and `label` as arguments, and returns a batch of loss values. Same interface as TensorOptimizer. ...
[ "Generic", "projected", "optimization", "generalized", "to", "work", "with", "approximate", "gradients", ".", "Used", "for", "e", ".", "g", ".", "the", "SPSA", "attack", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L613-L758
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
SPSA.generate
def generate(self, x, y=None, y_target=None, eps=None, clip_min=None, clip_max=None, nb_iter=None, is_targeted=None, early_stop_loss_threshold=None, learning_rate=DEFAULT...
python
def generate(self, x, y=None, y_target=None, eps=None, clip_min=None, clip_max=None, nb_iter=None, is_targeted=None, early_stop_loss_threshold=None, learning_rate=DEFAULT...
[ "def", "generate", "(", "self", ",", "x", ",", "y", "=", "None", ",", "y_target", "=", "None", ",", "eps", "=", "None", ",", "clip_min", "=", "None", ",", "clip_max", "=", "None", ",", "nb_iter", "=", "None", ",", "is_targeted", "=", "None", ",", ...
Generate symbolic graph for adversarial examples. :param x: The model's symbolic inputs. Must be a batch of size 1. :param y: A Tensor or None. The index of the correct label. :param y_target: A Tensor or None. The index of the target label in a targeted attack. :param eps: The siz...
[ "Generate", "symbolic", "graph", "for", "adversarial", "examples", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L51-L176
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
TensorOptimizer._compute_gradients
def _compute_gradients(self, loss_fn, x, unused_optim_state): """Compute a new value of `x` to minimize `loss_fn`. Args: loss_fn: a callable that takes `x`, a batch of images, and returns a batch of loss values. `x` will be optimized to minimize `loss_fn(x)`. x: A list o...
python
def _compute_gradients(self, loss_fn, x, unused_optim_state): """Compute a new value of `x` to minimize `loss_fn`. Args: loss_fn: a callable that takes `x`, a batch of images, and returns a batch of loss values. `x` will be optimized to minimize `loss_fn(x)`. x: A list o...
[ "def", "_compute_gradients", "(", "self", ",", "loss_fn", ",", "x", ",", "unused_optim_state", ")", ":", "# Assumes `x` is a list,", "# and contains a tensor representing a batch of images", "assert", "len", "(", "x", ")", "==", "1", "and", "isinstance", "(", "x", "...
Compute a new value of `x` to minimize `loss_fn`. Args: loss_fn: a callable that takes `x`, a batch of images, and returns a batch of loss values. `x` will be optimized to minimize `loss_fn(x)`. x: A list of Tensors, the values to be updated. This is analogous to...
[ "Compute", "a", "new", "value", "of", "x", "to", "minimize", "loss_fn", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L246-L270
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
TensorOptimizer.minimize
def minimize(self, loss_fn, x, optim_state): """ Analogous to tf.Optimizer.minimize :param loss_fn: tf Tensor, representing the loss to minimize :param x: list of Tensor, analogous to tf.Optimizer's var_list :param optim_state: A possibly nested dict, containing any optimizer state. Returns: ...
python
def minimize(self, loss_fn, x, optim_state): """ Analogous to tf.Optimizer.minimize :param loss_fn: tf Tensor, representing the loss to minimize :param x: list of Tensor, analogous to tf.Optimizer's var_list :param optim_state: A possibly nested dict, containing any optimizer state. Returns: ...
[ "def", "minimize", "(", "self", ",", "loss_fn", ",", "x", ",", "optim_state", ")", ":", "grads", "=", "self", ".", "_compute_gradients", "(", "loss_fn", ",", "x", ",", "optim_state", ")", "return", "self", ".", "_apply_gradients", "(", "grads", ",", "x",...
Analogous to tf.Optimizer.minimize :param loss_fn: tf Tensor, representing the loss to minimize :param x: list of Tensor, analogous to tf.Optimizer's var_list :param optim_state: A possibly nested dict, containing any optimizer state. Returns: new_x: list of Tensor, updated version of `x` ...
[ "Analogous", "to", "tf", ".", "Optimizer", ".", "minimize" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L287-L300
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
TensorAdam.init_state
def init_state(self, x): """ Initialize t, m, and u """ optim_state = {} optim_state["t"] = 0. optim_state["m"] = [tf.zeros_like(v) for v in x] optim_state["u"] = [tf.zeros_like(v) for v in x] return optim_state
python
def init_state(self, x): """ Initialize t, m, and u """ optim_state = {} optim_state["t"] = 0. optim_state["m"] = [tf.zeros_like(v) for v in x] optim_state["u"] = [tf.zeros_like(v) for v in x] return optim_state
[ "def", "init_state", "(", "self", ",", "x", ")", ":", "optim_state", "=", "{", "}", "optim_state", "[", "\"t\"", "]", "=", "0.", "optim_state", "[", "\"m\"", "]", "=", "[", "tf", ".", "zeros_like", "(", "v", ")", "for", "v", "in", "x", "]", "opti...
Initialize t, m, and u
[ "Initialize", "t", "m", "and", "u" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L340-L348
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
TensorAdam._apply_gradients
def _apply_gradients(self, grads, x, optim_state): """Refer to parent class documentation.""" new_x = [None] * len(x) new_optim_state = { "t": optim_state["t"] + 1., "m": [None] * len(x), "u": [None] * len(x) } t = new_optim_state["t"] for i in xrange(len(x)): g = g...
python
def _apply_gradients(self, grads, x, optim_state): """Refer to parent class documentation.""" new_x = [None] * len(x) new_optim_state = { "t": optim_state["t"] + 1., "m": [None] * len(x), "u": [None] * len(x) } t = new_optim_state["t"] for i in xrange(len(x)): g = g...
[ "def", "_apply_gradients", "(", "self", ",", "grads", ",", "x", ",", "optim_state", ")", ":", "new_x", "=", "[", "None", "]", "*", "len", "(", "x", ")", "new_optim_state", "=", "{", "\"t\"", ":", "optim_state", "[", "\"t\"", "]", "+", "1.", ",", "\...
Refer to parent class documentation.
[ "Refer", "to", "parent", "class", "documentation", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L350-L371
train
tensorflow/cleverhans
cleverhans/attacks/spsa.py
SPSAAdam._compute_gradients
def _compute_gradients(self, loss_fn, x, unused_optim_state): """Compute gradient estimates using SPSA.""" # Assumes `x` is a list, containing a [1, H, W, C] image # If static batch dimension is None, tf.reshape to batch size 1 # so that static shape can be inferred assert len(x) == 1 static_x_s...
python
def _compute_gradients(self, loss_fn, x, unused_optim_state): """Compute gradient estimates using SPSA.""" # Assumes `x` is a list, containing a [1, H, W, C] image # If static batch dimension is None, tf.reshape to batch size 1 # so that static shape can be inferred assert len(x) == 1 static_x_s...
[ "def", "_compute_gradients", "(", "self", ",", "loss_fn", ",", "x", ",", "unused_optim_state", ")", ":", "# Assumes `x` is a list, containing a [1, H, W, C] image", "# If static batch dimension is None, tf.reshape to batch size 1", "# so that static shape can be inferred", "assert", ...
Compute gradient estimates using SPSA.
[ "Compute", "gradient", "estimates", "using", "SPSA", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/spsa.py#L404-L441
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
parse_args
def parse_args(): """Parses command line arguments.""" parser = argparse.ArgumentParser( description='Tool to run attacks and defenses.') parser.add_argument('--attacks_dir', required=True, help='Location of all attacks.') parser.add_argument('--targeted_attacks_dir', required=True, ...
python
def parse_args(): """Parses command line arguments.""" parser = argparse.ArgumentParser( description='Tool to run attacks and defenses.') parser.add_argument('--attacks_dir', required=True, help='Location of all attacks.') parser.add_argument('--targeted_attacks_dir', required=True, ...
[ "def", "parse_args", "(", ")", ":", "parser", "=", "argparse", ".", "ArgumentParser", "(", "description", "=", "'Tool to run attacks and defenses.'", ")", "parser", ".", "add_argument", "(", "'--attacks_dir'", ",", "required", "=", "True", ",", "help", "=", "'Lo...
Parses command line arguments.
[ "Parses", "command", "line", "arguments", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L16-L44
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
read_submissions_from_directory
def read_submissions_from_directory(dirname, use_gpu): """Scans directory and read all submissions. Args: dirname: directory to scan. use_gpu: whether submissions should use GPU. This argument is used to pick proper Docker container for each submission and create instance of Attack or Defense c...
python
def read_submissions_from_directory(dirname, use_gpu): """Scans directory and read all submissions. Args: dirname: directory to scan. use_gpu: whether submissions should use GPU. This argument is used to pick proper Docker container for each submission and create instance of Attack or Defense c...
[ "def", "read_submissions_from_directory", "(", "dirname", ",", "use_gpu", ")", ":", "result", "=", "[", "]", "for", "sub_dir", "in", "os", ".", "listdir", "(", "dirname", ")", ":", "submission_path", "=", "os", ".", "path", ".", "join", "(", "dirname", "...
Scans directory and read all submissions. Args: dirname: directory to scan. use_gpu: whether submissions should use GPU. This argument is used to pick proper Docker container for each submission and create instance of Attack or Defense class. Returns: List with submissions (subclasses of S...
[ "Scans", "directory", "and", "read", "all", "submissions", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L121-L158
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
load_defense_output
def load_defense_output(filename): """Loads output of defense from given file.""" result = {} with open(filename) as f: for row in csv.reader(f): try: image_filename = row[0] if image_filename.endswith('.png') or image_filename.endswith('.jpg'): image_filename = image_filename[...
python
def load_defense_output(filename): """Loads output of defense from given file.""" result = {} with open(filename) as f: for row in csv.reader(f): try: image_filename = row[0] if image_filename.endswith('.png') or image_filename.endswith('.jpg'): image_filename = image_filename[...
[ "def", "load_defense_output", "(", "filename", ")", ":", "result", "=", "{", "}", "with", "open", "(", "filename", ")", "as", "f", ":", "for", "row", "in", "csv", ".", "reader", "(", "f", ")", ":", "try", ":", "image_filename", "=", "row", "[", "0"...
Loads output of defense from given file.
[ "Loads", "output", "of", "defense", "from", "given", "file", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L328-L341
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
compute_and_save_scores_and_ranking
def compute_and_save_scores_and_ranking(attacks_output, defenses_output, dataset_meta, output_dir, save_all_classification=False): """Computes scores and rank...
python
def compute_and_save_scores_and_ranking(attacks_output, defenses_output, dataset_meta, output_dir, save_all_classification=False): """Computes scores and rank...
[ "def", "compute_and_save_scores_and_ranking", "(", "attacks_output", ",", "defenses_output", ",", "dataset_meta", ",", "output_dir", ",", "save_all_classification", "=", "False", ")", ":", "def", "write_ranking", "(", "filename", ",", "header", ",", "names", ",", "s...
Computes scores and ranking and saves it. Args: attacks_output: output of attacks, instance of AttacksOutput class. defenses_output: outputs of defenses. Dictionary of dictionaries, key in outer dictionary is name of the defense, key of inner dictionary is name of the image, value of inner dictio...
[ "Computes", "scores", "and", "ranking", "and", "saves", "it", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L344-L465
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
main
def main(): """Run all attacks against all defenses and compute results. """ args = parse_args() attacks_output_dir = os.path.join(args.intermediate_results_dir, 'attacks_output') targeted_attacks_output_dir = os.path.join(args.intermediate_results_dir, ...
python
def main(): """Run all attacks against all defenses and compute results. """ args = parse_args() attacks_output_dir = os.path.join(args.intermediate_results_dir, 'attacks_output') targeted_attacks_output_dir = os.path.join(args.intermediate_results_dir, ...
[ "def", "main", "(", ")", ":", "args", "=", "parse_args", "(", ")", "attacks_output_dir", "=", "os", ".", "path", ".", "join", "(", "args", ".", "intermediate_results_dir", ",", "'attacks_output'", ")", "targeted_attacks_output_dir", "=", "os", ".", "path", "...
Run all attacks against all defenses and compute results.
[ "Run", "all", "attacks", "against", "all", "defenses", "and", "compute", "results", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L468-L547
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
Attack.run
def run(self, input_dir, output_dir, epsilon): """Runs attack inside Docker. Args: input_dir: directory with input (dataset). output_dir: directory where output (adversarial images) should be written. epsilon: maximum allowed size of adversarial perturbation, should be in range [0, 25...
python
def run(self, input_dir, output_dir, epsilon): """Runs attack inside Docker. Args: input_dir: directory with input (dataset). output_dir: directory where output (adversarial images) should be written. epsilon: maximum allowed size of adversarial perturbation, should be in range [0, 25...
[ "def", "run", "(", "self", ",", "input_dir", ",", "output_dir", ",", "epsilon", ")", ":", "print", "(", "'Running attack '", ",", "self", ".", "name", ")", "cmd", "=", "[", "self", ".", "docker_binary", "(", ")", ",", "'run'", ",", "'-v'", ",", "'{0}...
Runs attack inside Docker. Args: input_dir: directory with input (dataset). output_dir: directory where output (adversarial images) should be written. epsilon: maximum allowed size of adversarial perturbation, should be in range [0, 255].
[ "Runs", "attack", "inside", "Docker", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L73-L94
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
AttacksOutput._load_dataset_clipping
def _load_dataset_clipping(self, dataset_dir, epsilon): """Helper method which loads dataset and determines clipping range. Args: dataset_dir: location of the dataset. epsilon: maximum allowed size of adversarial perturbation. """ self.dataset_max_clip = {} self.dataset_min_clip = {} ...
python
def _load_dataset_clipping(self, dataset_dir, epsilon): """Helper method which loads dataset and determines clipping range. Args: dataset_dir: location of the dataset. epsilon: maximum allowed size of adversarial perturbation. """ self.dataset_max_clip = {} self.dataset_min_clip = {} ...
[ "def", "_load_dataset_clipping", "(", "self", ",", "dataset_dir", ",", "epsilon", ")", ":", "self", ".", "dataset_max_clip", "=", "{", "}", "self", ".", "dataset_min_clip", "=", "{", "}", "self", ".", "_dataset_image_count", "=", "0", "for", "fname", "in", ...
Helper method which loads dataset and determines clipping range. Args: dataset_dir: location of the dataset. epsilon: maximum allowed size of adversarial perturbation.
[ "Helper", "method", "which", "loads", "dataset", "and", "determines", "clipping", "range", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L191-L214
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
AttacksOutput.clip_and_copy_attack_outputs
def clip_and_copy_attack_outputs(self, attack_name, is_targeted): """Clips results of attack and copy it to directory with all images. Args: attack_name: name of the attack. is_targeted: if True then attack is targeted, otherwise non-targeted. """ if is_targeted: self._targeted_attack...
python
def clip_and_copy_attack_outputs(self, attack_name, is_targeted): """Clips results of attack and copy it to directory with all images. Args: attack_name: name of the attack. is_targeted: if True then attack is targeted, otherwise non-targeted. """ if is_targeted: self._targeted_attack...
[ "def", "clip_and_copy_attack_outputs", "(", "self", ",", "attack_name", ",", "is_targeted", ")", ":", "if", "is_targeted", ":", "self", ".", "_targeted_attack_names", ".", "add", "(", "attack_name", ")", "else", ":", "self", ".", "_attack_names", ".", "add", "...
Clips results of attack and copy it to directory with all images. Args: attack_name: name of the attack. is_targeted: if True then attack is targeted, otherwise non-targeted.
[ "Clips", "results", "of", "attack", "and", "copy", "it", "to", "directory", "with", "all", "images", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L216-L254
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py
DatasetMetadata.save_target_classes
def save_target_classes(self, filename): """Saves target classed for all dataset images into given file.""" with open(filename, 'w') as f: for k, v in self._target_classes.items(): f.write('{0}.png,{1}\n'.format(k, v))
python
def save_target_classes(self, filename): """Saves target classed for all dataset images into given file.""" with open(filename, 'w') as f: for k, v in self._target_classes.items(): f.write('{0}.png,{1}\n'.format(k, v))
[ "def", "save_target_classes", "(", "self", ",", "filename", ")", ":", "with", "open", "(", "filename", ",", "'w'", ")", "as", "f", ":", "for", "k", ",", "v", "in", "self", ".", "_target_classes", ".", "items", "(", ")", ":", "f", ".", "write", "(",...
Saves target classed for all dataset images into given file.
[ "Saves", "target", "classed", "for", "all", "dataset", "images", "into", "given", "file", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/run_attacks_and_defenses.py#L321-L325
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
single_run_max_confidence_recipe
def single_run_max_confidence_recipe(sess, model, x, y, nb_classes, eps, clip_min, clip_max, eps_iter, nb_iter, report_path, batch_size=BATCH_SIZE, eps_iter_small=None): ...
python
def single_run_max_confidence_recipe(sess, model, x, y, nb_classes, eps, clip_min, clip_max, eps_iter, nb_iter, report_path, batch_size=BATCH_SIZE, eps_iter_small=None): ...
[ "def", "single_run_max_confidence_recipe", "(", "sess", ",", "model", ",", "x", ",", "y", ",", "nb_classes", ",", "eps", ",", "clip_min", ",", "clip_max", ",", "eps_iter", ",", "nb_iter", ",", "report_path", ",", "batch_size", "=", "BATCH_SIZE", ",", "eps_it...
A reasonable attack bundling recipe for a max norm threat model and a defender that uses confidence thresholding. This recipe uses both uniform noise and randomly-initialized PGD targeted attacks. References: https://openreview.net/forum?id=H1g0piA9tQ This version runs each attack (noise, targeted PGD for e...
[ "A", "reasonable", "attack", "bundling", "recipe", "for", "a", "max", "norm", "threat", "model", "and", "a", "defender", "that", "uses", "confidence", "thresholding", ".", "This", "recipe", "uses", "both", "uniform", "noise", "and", "randomly", "-", "initializ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L43-L107
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
random_search_max_confidence_recipe
def random_search_max_confidence_recipe(sess, model, x, y, eps, clip_min, clip_max, report_path, batch_size=BATCH_SIZE, num_noise_points=10000): """Max confidence using random search. References:...
python
def random_search_max_confidence_recipe(sess, model, x, y, eps, clip_min, clip_max, report_path, batch_size=BATCH_SIZE, num_noise_points=10000): """Max confidence using random search. References:...
[ "def", "random_search_max_confidence_recipe", "(", "sess", ",", "model", ",", "x", ",", "y", ",", "eps", ",", "clip_min", ",", "clip_max", ",", "report_path", ",", "batch_size", "=", "BATCH_SIZE", ",", "num_noise_points", "=", "10000", ")", ":", "noise_attack"...
Max confidence using random search. References: https://openreview.net/forum?id=H1g0piA9tQ Describes the max_confidence procedure used for the bundling in this recipe https://arxiv.org/abs/1802.00420 Describes using random search with 1e5 or more random points to avoid gradient masking. :param ses...
[ "Max", "confidence", "using", "random", "search", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L256-L289
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
bundle_attacks
def bundle_attacks(sess, model, x, y, attack_configs, goals, report_path, attack_batch_size=BATCH_SIZE, eval_batch_size=BATCH_SIZE): """ Runs attack bundling. Users of cleverhans may call this function but are more likely to call one of the recipes above. Reference: https://openreview.net/...
python
def bundle_attacks(sess, model, x, y, attack_configs, goals, report_path, attack_batch_size=BATCH_SIZE, eval_batch_size=BATCH_SIZE): """ Runs attack bundling. Users of cleverhans may call this function but are more likely to call one of the recipes above. Reference: https://openreview.net/...
[ "def", "bundle_attacks", "(", "sess", ",", "model", ",", "x", ",", "y", ",", "attack_configs", ",", "goals", ",", "report_path", ",", "attack_batch_size", "=", "BATCH_SIZE", ",", "eval_batch_size", "=", "BATCH_SIZE", ")", ":", "assert", "isinstance", "(", "s...
Runs attack bundling. Users of cleverhans may call this function but are more likely to call one of the recipes above. Reference: https://openreview.net/forum?id=H1g0piA9tQ :param sess: tf.session.Session :param model: cleverhans.model.Model :param x: numpy array containing clean example inputs to attack ...
[ "Runs", "attack", "bundling", ".", "Users", "of", "cleverhans", "may", "call", "this", "function", "but", "are", "more", "likely", "to", "call", "one", "of", "the", "recipes", "above", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L320-L383
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
bundle_attacks_with_goal
def bundle_attacks_with_goal(sess, model, x, y, adv_x, attack_configs, run_counts, goal, report, report_path, attack_batch_size=BATCH_SIZE, eval_batch_size=BATCH_SIZE): """ Runs attack bundling, working on one specific AttackGoal...
python
def bundle_attacks_with_goal(sess, model, x, y, adv_x, attack_configs, run_counts, goal, report, report_path, attack_batch_size=BATCH_SIZE, eval_batch_size=BATCH_SIZE): """ Runs attack bundling, working on one specific AttackGoal...
[ "def", "bundle_attacks_with_goal", "(", "sess", ",", "model", ",", "x", ",", "y", ",", "adv_x", ",", "attack_configs", ",", "run_counts", ",", "goal", ",", "report", ",", "report_path", ",", "attack_batch_size", "=", "BATCH_SIZE", ",", "eval_batch_size", "=", ...
Runs attack bundling, working on one specific AttackGoal. This function is mostly intended to be called by `bundle_attacks`. Reference: https://openreview.net/forum?id=H1g0piA9tQ :param sess: tf.session.Session :param model: cleverhans.model.Model :param x: numpy array containing clean example inputs to att...
[ "Runs", "attack", "bundling", "working", "on", "one", "specific", "AttackGoal", ".", "This", "function", "is", "mostly", "intended", "to", "be", "called", "by", "bundle_attacks", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L385-L425
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
run_batch_with_goal
def run_batch_with_goal(sess, model, x, y, adv_x_val, criteria, attack_configs, run_counts, goal, report, report_path, attack_batch_size=BATCH_SIZE): """ Runs attack bundling on one batch of data. This function is mostly intended to be called by `bundle_attacks_wi...
python
def run_batch_with_goal(sess, model, x, y, adv_x_val, criteria, attack_configs, run_counts, goal, report, report_path, attack_batch_size=BATCH_SIZE): """ Runs attack bundling on one batch of data. This function is mostly intended to be called by `bundle_attacks_wi...
[ "def", "run_batch_with_goal", "(", "sess", ",", "model", ",", "x", ",", "y", ",", "adv_x_val", ",", "criteria", ",", "attack_configs", ",", "run_counts", ",", "goal", ",", "report", ",", "report_path", ",", "attack_batch_size", "=", "BATCH_SIZE", ")", ":", ...
Runs attack bundling on one batch of data. This function is mostly intended to be called by `bundle_attacks_with_goal`. :param sess: tf.session.Session :param model: cleverhans.model.Model :param x: numpy array containing clean example inputs to attack :param y: numpy array containing true labels :param ...
[ "Runs", "attack", "bundling", "on", "one", "batch", "of", "data", ".", "This", "function", "is", "mostly", "intended", "to", "be", "called", "by", "bundle_attacks_with_goal", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L428-L487
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
save
def save(criteria, report, report_path, adv_x_val): """ Saves the report and adversarial examples. :param criteria: dict, of the form returned by AttackGoal.get_criteria :param report: dict containing a confidence report :param report_path: string, filepath :param adv_x_val: numpy array containing dataset o...
python
def save(criteria, report, report_path, adv_x_val): """ Saves the report and adversarial examples. :param criteria: dict, of the form returned by AttackGoal.get_criteria :param report: dict containing a confidence report :param report_path: string, filepath :param adv_x_val: numpy array containing dataset o...
[ "def", "save", "(", "criteria", ",", "report", ",", "report_path", ",", "adv_x_val", ")", ":", "print_stats", "(", "criteria", "[", "'correctness'", "]", ",", "criteria", "[", "'confidence'", "]", ",", "'bundled'", ")", "print", "(", "\"Saving to \"", "+", ...
Saves the report and adversarial examples. :param criteria: dict, of the form returned by AttackGoal.get_criteria :param report: dict containing a confidence report :param report_path: string, filepath :param adv_x_val: numpy array containing dataset of adversarial examples
[ "Saves", "the", "report", "and", "adversarial", "examples", ".", ":", "param", "criteria", ":", "dict", "of", "the", "form", "returned", "by", "AttackGoal", ".", "get_criteria", ":", "param", "report", ":", "dict", "containing", "a", "confidence", "report", ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L490-L505
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
unfinished_attack_configs
def unfinished_attack_configs(new_work_goal, work_before, run_counts, log=False): """ Returns a list of attack configs that have not yet been run the desired number of times. :param new_work_goal: dict mapping attacks to desired number of times to run :param work_before: dict map...
python
def unfinished_attack_configs(new_work_goal, work_before, run_counts, log=False): """ Returns a list of attack configs that have not yet been run the desired number of times. :param new_work_goal: dict mapping attacks to desired number of times to run :param work_before: dict map...
[ "def", "unfinished_attack_configs", "(", "new_work_goal", ",", "work_before", ",", "run_counts", ",", "log", "=", "False", ")", ":", "assert", "isinstance", "(", "work_before", ",", "dict", ")", ",", "work_before", "for", "key", "in", "work_before", ":", "valu...
Returns a list of attack configs that have not yet been run the desired number of times. :param new_work_goal: dict mapping attacks to desired number of times to run :param work_before: dict mapping attacks to number of times they were run before starting this new goal. Should be prefiltered to include only ...
[ "Returns", "a", "list", "of", "attack", "configs", "that", "have", "not", "yet", "been", "run", "the", "desired", "number", "of", "times", ".", ":", "param", "new_work_goal", ":", "dict", "mapping", "attacks", "to", "desired", "number", "of", "times", "to"...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L916-L962
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
bundle_examples_with_goal
def bundle_examples_with_goal(sess, model, adv_x_list, y, goal, report_path, batch_size=BATCH_SIZE): """ A post-processor version of attack bundling, that chooses the strongest example from the output of multiple earlier bundling strategies. :param sess: tf.session.Session :para...
python
def bundle_examples_with_goal(sess, model, adv_x_list, y, goal, report_path, batch_size=BATCH_SIZE): """ A post-processor version of attack bundling, that chooses the strongest example from the output of multiple earlier bundling strategies. :param sess: tf.session.Session :para...
[ "def", "bundle_examples_with_goal", "(", "sess", ",", "model", ",", "adv_x_list", ",", "y", ",", "goal", ",", "report_path", ",", "batch_size", "=", "BATCH_SIZE", ")", ":", "# Check the input", "num_attacks", "=", "len", "(", "adv_x_list", ")", "assert", "num_...
A post-processor version of attack bundling, that chooses the strongest example from the output of multiple earlier bundling strategies. :param sess: tf.session.Session :param model: cleverhans.model.Model :param adv_x_list: list of numpy arrays Each entry in the list is the output of a previous bundler; i...
[ "A", "post", "-", "processor", "version", "of", "attack", "bundling", "that", "chooses", "the", "strongest", "example", "from", "the", "output", "of", "multiple", "earlier", "bundling", "strategies", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L1044-L1110
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
spsa_max_confidence_recipe
def spsa_max_confidence_recipe(sess, model, x, y, nb_classes, eps, clip_min, clip_max, nb_iter, report_path, spsa_samples=SPSA.DEFAULT_SPSA_SAMPLES, spsa_iters=SPSA.DEFAULT_SPSA_ITERS, ...
python
def spsa_max_confidence_recipe(sess, model, x, y, nb_classes, eps, clip_min, clip_max, nb_iter, report_path, spsa_samples=SPSA.DEFAULT_SPSA_SAMPLES, spsa_iters=SPSA.DEFAULT_SPSA_ITERS, ...
[ "def", "spsa_max_confidence_recipe", "(", "sess", ",", "model", ",", "x", ",", "y", ",", "nb_classes", ",", "eps", ",", "clip_min", ",", "clip_max", ",", "nb_iter", ",", "report_path", ",", "spsa_samples", "=", "SPSA", ".", "DEFAULT_SPSA_SAMPLES", ",", "spsa...
Runs the MaxConfidence attack using SPSA as the underlying optimizer. Even though this runs only one attack, it must be implemented as a bundler because SPSA supports only batch_size=1. The cleverhans.attacks.MaxConfidence attack internally multiplies the batch size by nb_classes, so it can't take SPSA as a ba...
[ "Runs", "the", "MaxConfidence", "attack", "using", "SPSA", "as", "the", "underlying", "optimizer", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L1112-L1156
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
AttackGoal.get_criteria
def get_criteria(self, sess, model, advx, y, batch_size=BATCH_SIZE): """ Returns a dictionary mapping the name of each criterion to a NumPy array containing the value of that criterion for each adversarial example. Subclasses can add extra criteria by implementing the `extra_criteria` method. ...
python
def get_criteria(self, sess, model, advx, y, batch_size=BATCH_SIZE): """ Returns a dictionary mapping the name of each criterion to a NumPy array containing the value of that criterion for each adversarial example. Subclasses can add extra criteria by implementing the `extra_criteria` method. ...
[ "def", "get_criteria", "(", "self", ",", "sess", ",", "model", ",", "advx", ",", "y", ",", "batch_size", "=", "BATCH_SIZE", ")", ":", "names", ",", "factory", "=", "self", ".", "extra_criteria", "(", ")", "factory", "=", "_CriteriaFactory", "(", "model",...
Returns a dictionary mapping the name of each criterion to a NumPy array containing the value of that criterion for each adversarial example. Subclasses can add extra criteria by implementing the `extra_criteria` method. :param sess: tf.session.Session :param model: cleverhans.model.Model :...
[ "Returns", "a", "dictionary", "mapping", "the", "name", "of", "each", "criterion", "to", "a", "NumPy", "array", "containing", "the", "value", "of", "that", "criterion", "for", "each", "adversarial", "example", ".", "Subclasses", "can", "add", "extra", "criteri...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L532-L554
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
AttackGoal.request_examples
def request_examples(self, attack_config, criteria, run_counts, batch_size): """ Returns a numpy array of integer example indices to run in the next batch. """ raise NotImplementedError(str(type(self)) + "needs to implement request_examples")
python
def request_examples(self, attack_config, criteria, run_counts, batch_size): """ Returns a numpy array of integer example indices to run in the next batch. """ raise NotImplementedError(str(type(self)) + "needs to implement request_examples")
[ "def", "request_examples", "(", "self", ",", "attack_config", ",", "criteria", ",", "run_counts", ",", "batch_size", ")", ":", "raise", "NotImplementedError", "(", "str", "(", "type", "(", "self", ")", ")", "+", "\"needs to implement request_examples\"", ")" ]
Returns a numpy array of integer example indices to run in the next batch.
[ "Returns", "a", "numpy", "array", "of", "integer", "example", "indices", "to", "run", "in", "the", "next", "batch", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L563-L568
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
AttackGoal.new_wins
def new_wins(self, orig_criteria, orig_idx, new_criteria, new_idx): """ Returns a bool indicating whether a new adversarial example is better than the pre-existing one for the same clean example. :param orig_criteria: dict mapping names of criteria to their value for each example in the whole data...
python
def new_wins(self, orig_criteria, orig_idx, new_criteria, new_idx): """ Returns a bool indicating whether a new adversarial example is better than the pre-existing one for the same clean example. :param orig_criteria: dict mapping names of criteria to their value for each example in the whole data...
[ "def", "new_wins", "(", "self", ",", "orig_criteria", ",", "orig_idx", ",", "new_criteria", ",", "new_idx", ")", ":", "raise", "NotImplementedError", "(", "str", "(", "type", "(", "self", ")", ")", "+", "\" needs to implement new_wins.\"", ")" ]
Returns a bool indicating whether a new adversarial example is better than the pre-existing one for the same clean example. :param orig_criteria: dict mapping names of criteria to their value for each example in the whole dataset :param orig_idx: The position of the pre-existing example within the ...
[ "Returns", "a", "bool", "indicating", "whether", "a", "new", "adversarial", "example", "is", "better", "than", "the", "pre", "-", "existing", "one", "for", "the", "same", "clean", "example", ".", ":", "param", "orig_criteria", ":", "dict", "mapping", "names"...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L593-L607
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
Misclassify.filter
def filter(self, run_counts, criteria): """ Return run counts only for examples that are still correctly classified """ correctness = criteria['correctness'] assert correctness.dtype == np.bool filtered_counts = deep_copy(run_counts) for key in filtered_counts: filtered_counts[key] = f...
python
def filter(self, run_counts, criteria): """ Return run counts only for examples that are still correctly classified """ correctness = criteria['correctness'] assert correctness.dtype == np.bool filtered_counts = deep_copy(run_counts) for key in filtered_counts: filtered_counts[key] = f...
[ "def", "filter", "(", "self", ",", "run_counts", ",", "criteria", ")", ":", "correctness", "=", "criteria", "[", "'correctness'", "]", "assert", "correctness", ".", "dtype", "==", "np", ".", "bool", "filtered_counts", "=", "deep_copy", "(", "run_counts", ")"...
Return run counts only for examples that are still correctly classified
[ "Return", "run", "counts", "only", "for", "examples", "that", "are", "still", "correctly", "classified" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L690-L699
train
tensorflow/cleverhans
cleverhans/attack_bundling.py
MaxConfidence.filter
def filter(self, run_counts, criteria): """ Return the counts for only those examples that are below the threshold """ wrong_confidence = criteria['wrong_confidence'] below_t = wrong_confidence <= self.t filtered_counts = deep_copy(run_counts) for key in filtered_counts: filtered_count...
python
def filter(self, run_counts, criteria): """ Return the counts for only those examples that are below the threshold """ wrong_confidence = criteria['wrong_confidence'] below_t = wrong_confidence <= self.t filtered_counts = deep_copy(run_counts) for key in filtered_counts: filtered_count...
[ "def", "filter", "(", "self", ",", "run_counts", ",", "criteria", ")", ":", "wrong_confidence", "=", "criteria", "[", "'wrong_confidence'", "]", "below_t", "=", "wrong_confidence", "<=", "self", ".", "t", "filtered_counts", "=", "deep_copy", "(", "run_counts", ...
Return the counts for only those examples that are below the threshold
[ "Return", "the", "counts", "for", "only", "those", "examples", "that", "are", "below", "the", "threshold" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attack_bundling.py#L799-L808
train
tensorflow/cleverhans
cleverhans/future/tf2/attacks/projected_gradient_descent.py
projected_gradient_descent
def projected_gradient_descent(model_fn, x, eps, eps_iter, nb_iter, ord, clip_min=None, clip_max=None, y=None, targeted=False, rand_init=None, rand_minmax=0.3, sanity_checks=True): """ This class implements either the Basic Iterative Method (Kurakin et...
python
def projected_gradient_descent(model_fn, x, eps, eps_iter, nb_iter, ord, clip_min=None, clip_max=None, y=None, targeted=False, rand_init=None, rand_minmax=0.3, sanity_checks=True): """ This class implements either the Basic Iterative Method (Kurakin et...
[ "def", "projected_gradient_descent", "(", "model_fn", ",", "x", ",", "eps", ",", "eps_iter", ",", "nb_iter", ",", "ord", ",", "clip_min", "=", "None", ",", "clip_max", "=", "None", ",", "y", "=", "None", ",", "targeted", "=", "False", ",", "rand_init", ...
This class implements either the Basic Iterative Method (Kurakin et al. 2016) when rand_init is set to 0. or the Madry et al. (2017) method when rand_minmax is larger than 0. Paper link (Kurakin et al. 2016): https://arxiv.org/pdf/1607.02533.pdf Paper link (Madry et al. 2017): https://arxiv.org/pdf/1706.06083.p...
[ "This", "class", "implements", "either", "the", "Basic", "Iterative", "Method", "(", "Kurakin", "et", "al", ".", "2016", ")", "when", "rand_init", "is", "set", "to", "0", ".", "or", "the", "Madry", "et", "al", ".", "(", "2017", ")", "method", "when", ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/future/tf2/attacks/projected_gradient_descent.py#L10-L100
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
clip_image
def clip_image(image, clip_min, clip_max): """ Clip an image, or an image batch, with upper and lower threshold. """ return np.minimum(np.maximum(clip_min, image), clip_max)
python
def clip_image(image, clip_min, clip_max): """ Clip an image, or an image batch, with upper and lower threshold. """ return np.minimum(np.maximum(clip_min, image), clip_max)
[ "def", "clip_image", "(", "image", ",", "clip_min", ",", "clip_max", ")", ":", "return", "np", ".", "minimum", "(", "np", ".", "maximum", "(", "clip_min", ",", "image", ")", ",", "clip_max", ")" ]
Clip an image, or an image batch, with upper and lower threshold.
[ "Clip", "an", "image", "or", "an", "image", "batch", "with", "upper", "and", "lower", "threshold", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L353-L355
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
compute_distance
def compute_distance(x_ori, x_pert, constraint='l2'): """ Compute the distance between two images. """ if constraint == 'l2': dist = np.linalg.norm(x_ori - x_pert) elif constraint == 'linf': dist = np.max(abs(x_ori - x_pert)) return dist
python
def compute_distance(x_ori, x_pert, constraint='l2'): """ Compute the distance between two images. """ if constraint == 'l2': dist = np.linalg.norm(x_ori - x_pert) elif constraint == 'linf': dist = np.max(abs(x_ori - x_pert)) return dist
[ "def", "compute_distance", "(", "x_ori", ",", "x_pert", ",", "constraint", "=", "'l2'", ")", ":", "if", "constraint", "==", "'l2'", ":", "dist", "=", "np", ".", "linalg", ".", "norm", "(", "x_ori", "-", "x_pert", ")", "elif", "constraint", "==", "'linf...
Compute the distance between two images.
[ "Compute", "the", "distance", "between", "two", "images", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L358-L364
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
approximate_gradient
def approximate_gradient(decision_function, sample, num_evals, delta, constraint, shape, clip_min, clip_max): """ Gradient direction estimation """ # Generate random vectors. noise_shape = [num_evals] + list(shape) if constraint == 'l2': rv = np.random.randn(*noise_shape) elif con...
python
def approximate_gradient(decision_function, sample, num_evals, delta, constraint, shape, clip_min, clip_max): """ Gradient direction estimation """ # Generate random vectors. noise_shape = [num_evals] + list(shape) if constraint == 'l2': rv = np.random.randn(*noise_shape) elif con...
[ "def", "approximate_gradient", "(", "decision_function", ",", "sample", ",", "num_evals", ",", "delta", ",", "constraint", ",", "shape", ",", "clip_min", ",", "clip_max", ")", ":", "# Generate random vectors.", "noise_shape", "=", "[", "num_evals", "]", "+", "li...
Gradient direction estimation
[ "Gradient", "direction", "estimation" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L366-L399
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
project
def project(original_image, perturbed_images, alphas, shape, constraint): """ Projection onto given l2 / linf balls in a batch. """ alphas_shape = [len(alphas)] + [1] * len(shape) alphas = alphas.reshape(alphas_shape) if constraint == 'l2': projected = (1-alphas) * original_image + alphas * perturbed_images...
python
def project(original_image, perturbed_images, alphas, shape, constraint): """ Projection onto given l2 / linf balls in a batch. """ alphas_shape = [len(alphas)] + [1] * len(shape) alphas = alphas.reshape(alphas_shape) if constraint == 'l2': projected = (1-alphas) * original_image + alphas * perturbed_images...
[ "def", "project", "(", "original_image", ",", "perturbed_images", ",", "alphas", ",", "shape", ",", "constraint", ")", ":", "alphas_shape", "=", "[", "len", "(", "alphas", ")", "]", "+", "[", "1", "]", "*", "len", "(", "shape", ")", "alphas", "=", "a...
Projection onto given l2 / linf balls in a batch.
[ "Projection", "onto", "given", "l2", "/", "linf", "balls", "in", "a", "batch", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L402-L414
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
binary_search_batch
def binary_search_batch(original_image, perturbed_images, decision_function, shape, constraint, theta): """ Binary search to approach the boundary. """ # Compute distance between each of perturbed image and original image. dists_post_update = np.array([ compute_distance( o...
python
def binary_search_batch(original_image, perturbed_images, decision_function, shape, constraint, theta): """ Binary search to approach the boundary. """ # Compute distance between each of perturbed image and original image. dists_post_update = np.array([ compute_distance( o...
[ "def", "binary_search_batch", "(", "original_image", ",", "perturbed_images", ",", "decision_function", ",", "shape", ",", "constraint", ",", "theta", ")", ":", "# Compute distance between each of perturbed image and original image.", "dists_post_update", "=", "np", ".", "a...
Binary search to approach the boundary.
[ "Binary", "search", "to", "approach", "the", "boundary", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L417-L468
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
initialize
def initialize(decision_function, sample, shape, clip_min, clip_max): """ Efficient Implementation of BlendedUniformNoiseAttack in Foolbox. """ success = 0 num_evals = 0 # Find a misclassified random noise. while True: random_noise = np.random.uniform(clip_min, clip_max, size=shape) success = dec...
python
def initialize(decision_function, sample, shape, clip_min, clip_max): """ Efficient Implementation of BlendedUniformNoiseAttack in Foolbox. """ success = 0 num_evals = 0 # Find a misclassified random noise. while True: random_noise = np.random.uniform(clip_min, clip_max, size=shape) success = dec...
[ "def", "initialize", "(", "decision_function", ",", "sample", ",", "shape", ",", "clip_min", ",", "clip_max", ")", ":", "success", "=", "0", "num_evals", "=", "0", "# Find a misclassified random noise.", "while", "True", ":", "random_noise", "=", "np", ".", "r...
Efficient Implementation of BlendedUniformNoiseAttack in Foolbox.
[ "Efficient", "Implementation", "of", "BlendedUniformNoiseAttack", "in", "Foolbox", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L471-L501
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
geometric_progression_for_stepsize
def geometric_progression_for_stepsize(x, update, dist, decision_function, current_iteration): """ Geometric progression to search for stepsize. Keep decreasing stepsize by half until reaching the desired side of the boundary. """ epsilon = dist / np.sqrt(current...
python
def geometric_progression_for_stepsize(x, update, dist, decision_function, current_iteration): """ Geometric progression to search for stepsize. Keep decreasing stepsize by half until reaching the desired side of the boundary. """ epsilon = dist / np.sqrt(current...
[ "def", "geometric_progression_for_stepsize", "(", "x", ",", "update", ",", "dist", ",", "decision_function", ",", "current_iteration", ")", ":", "epsilon", "=", "dist", "/", "np", ".", "sqrt", "(", "current_iteration", ")", "while", "True", ":", "updated", "="...
Geometric progression to search for stepsize. Keep decreasing stepsize by half until reaching the desired side of the boundary.
[ "Geometric", "progression", "to", "search", "for", "stepsize", ".", "Keep", "decreasing", "stepsize", "by", "half", "until", "reaching", "the", "desired", "side", "of", "the", "boundary", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L504-L519
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
select_delta
def select_delta(dist_post_update, current_iteration, clip_max, clip_min, d, theta, constraint): """ Choose the delta at the scale of distance between x and perturbed sample. """ if current_iteration == 1: delta = 0.1 * (clip_max - clip_min) else: if constraint == 'l2': delta...
python
def select_delta(dist_post_update, current_iteration, clip_max, clip_min, d, theta, constraint): """ Choose the delta at the scale of distance between x and perturbed sample. """ if current_iteration == 1: delta = 0.1 * (clip_max - clip_min) else: if constraint == 'l2': delta...
[ "def", "select_delta", "(", "dist_post_update", ",", "current_iteration", ",", "clip_max", ",", "clip_min", ",", "d", ",", "theta", ",", "constraint", ")", ":", "if", "current_iteration", "==", "1", ":", "delta", "=", "0.1", "*", "(", "clip_max", "-", "cli...
Choose the delta at the scale of distance between x and perturbed sample.
[ "Choose", "the", "delta", "at", "the", "scale", "of", "distance", "between", "x", "and", "perturbed", "sample", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L522-L536
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
BoundaryAttackPlusPlus.generate
def generate(self, x, **kwargs): """ Return a tensor that constructs adversarial examples for the given input. Generate uses tf.py_func in order to operate over tensors. :param x: A tensor with the inputs. :param kwargs: See `parse_params` """ self.parse_params(**kwargs) shape = [int(i) ...
python
def generate(self, x, **kwargs): """ Return a tensor that constructs adversarial examples for the given input. Generate uses tf.py_func in order to operate over tensors. :param x: A tensor with the inputs. :param kwargs: See `parse_params` """ self.parse_params(**kwargs) shape = [int(i) ...
[ "def", "generate", "(", "self", ",", "x", ",", "*", "*", "kwargs", ")", ":", "self", ".", "parse_params", "(", "*", "*", "kwargs", ")", "shape", "=", "[", "int", "(", "i", ")", "for", "i", "in", "x", ".", "get_shape", "(", ")", ".", "as_list", ...
Return a tensor that constructs adversarial examples for the given input. Generate uses tf.py_func in order to operate over tensors. :param x: A tensor with the inputs. :param kwargs: See `parse_params`
[ "Return", "a", "tensor", "that", "constructs", "adversarial", "examples", "for", "the", "given", "input", ".", "Generate", "uses", "tf", ".", "py_func", "in", "order", "to", "operate", "over", "tensors", ".", ":", "param", "x", ":", "A", "tensor", "with", ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L61-L120
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
BoundaryAttackPlusPlus.generate_np
def generate_np(self, x, **kwargs): """ Generate adversarial images in a for loop. :param y: An array of shape (n, nb_classes) for true labels. :param y_target: An array of shape (n, nb_classes) for target labels. Required for targeted attack. :param image_target: An array of shape (n, **image ...
python
def generate_np(self, x, **kwargs): """ Generate adversarial images in a for loop. :param y: An array of shape (n, nb_classes) for true labels. :param y_target: An array of shape (n, nb_classes) for target labels. Required for targeted attack. :param image_target: An array of shape (n, **image ...
[ "def", "generate_np", "(", "self", ",", "x", ",", "*", "*", "kwargs", ")", ":", "x_adv", "=", "[", "]", "if", "'image_target'", "in", "kwargs", "and", "kwargs", "[", "'image_target'", "]", "is", "not", "None", ":", "image_target", "=", "np", ".", "co...
Generate adversarial images in a for loop. :param y: An array of shape (n, nb_classes) for true labels. :param y_target: An array of shape (n, nb_classes) for target labels. Required for targeted attack. :param image_target: An array of shape (n, **image shape) for initial target images. Required f...
[ "Generate", "adversarial", "images", "in", "a", "for", "loop", ".", ":", "param", "y", ":", "An", "array", "of", "shape", "(", "n", "nb_classes", ")", "for", "true", "labels", ".", ":", "param", "y_target", ":", "An", "array", "of", "shape", "(", "n"...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L122-L159
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
BoundaryAttackPlusPlus.parse_params
def parse_params(self, y_target=None, image_target=None, initial_num_evals=100, max_num_evals=10000, stepsize_search='grid_search', num_iterations=64, gamma=0.01, const...
python
def parse_params(self, y_target=None, image_target=None, initial_num_evals=100, max_num_evals=10000, stepsize_search='grid_search', num_iterations=64, gamma=0.01, const...
[ "def", "parse_params", "(", "self", ",", "y_target", "=", "None", ",", "image_target", "=", "None", ",", "initial_num_evals", "=", "100", ",", "max_num_evals", "=", "10000", ",", "stepsize_search", "=", "'grid_search'", ",", "num_iterations", "=", "64", ",", ...
:param y: A tensor of shape (1, nb_classes) for true labels. :param y_target: A tensor of shape (1, nb_classes) for target labels. Required for targeted attack. :param image_target: A tensor of shape (1, **image shape) for initial target images. Required for targeted attack. :param initial_num_eval...
[ ":", "param", "y", ":", "A", "tensor", "of", "shape", "(", "1", "nb_classes", ")", "for", "true", "labels", ".", ":", "param", "y_target", ":", "A", "tensor", "of", "shape", "(", "1", "nb_classes", ")", "for", "target", "labels", ".", "Required", "fo...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L161-L213
train
tensorflow/cleverhans
cleverhans/attacks/bapp.py
BoundaryAttackPlusPlus._bapp
def _bapp(self, sample, target_label, target_image): """ Main algorithm for Boundary Attack ++. Return a tensor that constructs adversarial examples for the given input. Generate uses tf.py_func in order to operate over tensors. :param sample: input image. Without the batchsize dimension. :par...
python
def _bapp(self, sample, target_label, target_image): """ Main algorithm for Boundary Attack ++. Return a tensor that constructs adversarial examples for the given input. Generate uses tf.py_func in order to operate over tensors. :param sample: input image. Without the batchsize dimension. :par...
[ "def", "_bapp", "(", "self", ",", "sample", ",", "target_label", ",", "target_image", ")", ":", "# Original label required for untargeted attack.", "if", "target_label", "is", "None", ":", "original_label", "=", "np", ".", "argmax", "(", "self", ".", "sess", "."...
Main algorithm for Boundary Attack ++. Return a tensor that constructs adversarial examples for the given input. Generate uses tf.py_func in order to operate over tensors. :param sample: input image. Without the batchsize dimension. :param target_label: integer for targeted attack, None for nont...
[ "Main", "algorithm", "for", "Boundary", "Attack", "++", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/bapp.py#L215-L339
train
tensorflow/cleverhans
cleverhans/attacks/fast_feature_adversaries.py
FastFeatureAdversaries.parse_params
def parse_params(self, layer=None, eps=0.3, eps_iter=0.05, nb_iter=10, ord=np.inf, clip_min=None, clip_max=None, **kwargs): """ Take in a dictionary of paramete...
python
def parse_params(self, layer=None, eps=0.3, eps_iter=0.05, nb_iter=10, ord=np.inf, clip_min=None, clip_max=None, **kwargs): """ Take in a dictionary of paramete...
[ "def", "parse_params", "(", "self", ",", "layer", "=", "None", ",", "eps", "=", "0.3", ",", "eps_iter", "=", "0.05", ",", "nb_iter", "=", "10", ",", "ord", "=", "np", ".", "inf", ",", "clip_min", "=", "None", ",", "clip_max", "=", "None", ",", "*...
Take in a dictionary of parameters and applies attack-specific checks before saving them as attributes. Attack-specific parameters: :param layer: (required str) name of the layer to target. :param eps: (optional float) maximum distortion of adversarial example compared to original inpu...
[ "Take", "in", "a", "dictionary", "of", "parameters", "and", "applies", "attack", "-", "specific", "checks", "before", "saving", "them", "as", "attributes", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/fast_feature_adversaries.py#L44-L86
train
tensorflow/cleverhans
cleverhans/attacks/fast_feature_adversaries.py
FastFeatureAdversaries.attack_single_step
def attack_single_step(self, x, eta, g_feat): """ TensorFlow implementation of the Fast Feature Gradient. This is a single step attack similar to Fast Gradient Method that attacks an internal representation. :param x: the input placeholder :param eta: A tensor the same shape as x that holds the...
python
def attack_single_step(self, x, eta, g_feat): """ TensorFlow implementation of the Fast Feature Gradient. This is a single step attack similar to Fast Gradient Method that attacks an internal representation. :param x: the input placeholder :param eta: A tensor the same shape as x that holds the...
[ "def", "attack_single_step", "(", "self", ",", "x", ",", "eta", ",", "g_feat", ")", ":", "adv_x", "=", "x", "+", "eta", "a_feat", "=", "self", ".", "model", ".", "fprop", "(", "adv_x", ")", "[", "self", ".", "layer", "]", "# feat.shape = (batch, c) or ...
TensorFlow implementation of the Fast Feature Gradient. This is a single step attack similar to Fast Gradient Method that attacks an internal representation. :param x: the input placeholder :param eta: A tensor the same shape as x that holds the perturbation. :param g_feat: model's internal tensor ...
[ "TensorFlow", "implementation", "of", "the", "Fast", "Feature", "Gradient", ".", "This", "is", "a", "single", "step", "attack", "similar", "to", "Fast", "Gradient", "Method", "that", "attacks", "an", "internal", "representation", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/fast_feature_adversaries.py#L88-L129
train
tensorflow/cleverhans
cleverhans/attacks/fast_feature_adversaries.py
FastFeatureAdversaries.generate
def generate(self, x, g, **kwargs): """ Generate symbolic graph for adversarial examples and return. :param x: The model's symbolic inputs. :param g: The target value of the symbolic representation :param kwargs: See `parse_params` """ # Parse and save attack-specific parameters assert...
python
def generate(self, x, g, **kwargs): """ Generate symbolic graph for adversarial examples and return. :param x: The model's symbolic inputs. :param g: The target value of the symbolic representation :param kwargs: See `parse_params` """ # Parse and save attack-specific parameters assert...
[ "def", "generate", "(", "self", ",", "x", ",", "g", ",", "*", "*", "kwargs", ")", ":", "# Parse and save attack-specific parameters", "assert", "self", ".", "parse_params", "(", "*", "*", "kwargs", ")", "g_feat", "=", "self", ".", "model", ".", "fprop", ...
Generate symbolic graph for adversarial examples and return. :param x: The model's symbolic inputs. :param g: The target value of the symbolic representation :param kwargs: See `parse_params`
[ "Generate", "symbolic", "graph", "for", "adversarial", "examples", "and", "return", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/fast_feature_adversaries.py#L131-L165
train
tensorflow/cleverhans
scripts/make_confidence_report_bundled.py
main
def main(argv=None): """ Make a confidence report and save it to disk. """ try: _name_of_script, filepath = argv except ValueError: raise ValueError(argv) print(filepath) make_confidence_report_bundled(filepath=filepath, test_start=FLAGS.test_start, ...
python
def main(argv=None): """ Make a confidence report and save it to disk. """ try: _name_of_script, filepath = argv except ValueError: raise ValueError(argv) print(filepath) make_confidence_report_bundled(filepath=filepath, test_start=FLAGS.test_start, ...
[ "def", "main", "(", "argv", "=", "None", ")", ":", "try", ":", "_name_of_script", ",", "filepath", "=", "argv", "except", "ValueError", ":", "raise", "ValueError", "(", "argv", ")", "print", "(", "filepath", ")", "make_confidence_report_bundled", "(", "filep...
Make a confidence report and save it to disk.
[ "Make", "a", "confidence", "report", "and", "save", "it", "to", "disk", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/scripts/make_confidence_report_bundled.py#L42-L56
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py
block35
def block35(net, scale=1.0, activation_fn=tf.nn.relu, scope=None, reuse=None): """Builds the 35x35 resnet block.""" with tf.variable_scope(scope, 'Block35', [net], reuse=reuse): with tf.variable_scope('Branch_0'): tower_conv = slim.conv2d(net, 32, 1, scope='Conv2d_1x1') with tf.variable_scope('Branch_...
python
def block35(net, scale=1.0, activation_fn=tf.nn.relu, scope=None, reuse=None): """Builds the 35x35 resnet block.""" with tf.variable_scope(scope, 'Block35', [net], reuse=reuse): with tf.variable_scope('Branch_0'): tower_conv = slim.conv2d(net, 32, 1, scope='Conv2d_1x1') with tf.variable_scope('Branch_...
[ "def", "block35", "(", "net", ",", "scale", "=", "1.0", ",", "activation_fn", "=", "tf", ".", "nn", ".", "relu", ",", "scope", "=", "None", ",", "reuse", "=", "None", ")", ":", "with", "tf", ".", "variable_scope", "(", "scope", ",", "'Block35'", ",...
Builds the 35x35 resnet block.
[ "Builds", "the", "35x35", "resnet", "block", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py#L35-L54
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py
block17
def block17(net, scale=1.0, activation_fn=tf.nn.relu, scope=None, reuse=None): """Builds the 17x17 resnet block.""" with tf.variable_scope(scope, 'Block17', [net], reuse=reuse): with tf.variable_scope('Branch_0'): tower_conv = slim.conv2d(net, 192, 1, scope='Conv2d_1x1') with tf.variable_scope('Branch...
python
def block17(net, scale=1.0, activation_fn=tf.nn.relu, scope=None, reuse=None): """Builds the 17x17 resnet block.""" with tf.variable_scope(scope, 'Block17', [net], reuse=reuse): with tf.variable_scope('Branch_0'): tower_conv = slim.conv2d(net, 192, 1, scope='Conv2d_1x1') with tf.variable_scope('Branch...
[ "def", "block17", "(", "net", ",", "scale", "=", "1.0", ",", "activation_fn", "=", "tf", ".", "nn", ".", "relu", ",", "scope", "=", "None", ",", "reuse", "=", "None", ")", ":", "with", "tf", ".", "variable_scope", "(", "scope", ",", "'Block17'", ",...
Builds the 17x17 resnet block.
[ "Builds", "the", "17x17", "resnet", "block", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py#L57-L74
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py
inception_resnet_v2_base
def inception_resnet_v2_base(inputs, final_endpoint='Conv2d_7b_1x1', output_stride=16, align_feature_maps=False, scope=None): """Inception model from http://arxiv.org/abs/1602.07261. Constructs an I...
python
def inception_resnet_v2_base(inputs, final_endpoint='Conv2d_7b_1x1', output_stride=16, align_feature_maps=False, scope=None): """Inception model from http://arxiv.org/abs/1602.07261. Constructs an I...
[ "def", "inception_resnet_v2_base", "(", "inputs", ",", "final_endpoint", "=", "'Conv2d_7b_1x1'", ",", "output_stride", "=", "16", ",", "align_feature_maps", "=", "False", ",", "scope", "=", "None", ")", ":", "if", "output_stride", "!=", "8", "and", "output_strid...
Inception model from http://arxiv.org/abs/1602.07261. Constructs an Inception Resnet v2 network from inputs to the given final endpoint. This method can construct the network up to the final inception block Conv2d_7b_1x1. Args: inputs: a tensor of size [batch_size, height, width, channels]. final_end...
[ "Inception", "model", "from", "http", ":", "//", "arxiv", ".", "org", "/", "abs", "/", "1602", ".", "07261", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py#L97-L285
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py
inception_resnet_v2
def inception_resnet_v2(inputs, nb_classes=1001, is_training=True, dropout_keep_prob=0.8, reuse=None, scope='InceptionResnetV2', create_aux_logits=True, num_classes=None): """Creates the Inception R...
python
def inception_resnet_v2(inputs, nb_classes=1001, is_training=True, dropout_keep_prob=0.8, reuse=None, scope='InceptionResnetV2', create_aux_logits=True, num_classes=None): """Creates the Inception R...
[ "def", "inception_resnet_v2", "(", "inputs", ",", "nb_classes", "=", "1001", ",", "is_training", "=", "True", ",", "dropout_keep_prob", "=", "0.8", ",", "reuse", "=", "None", ",", "scope", "=", "'InceptionResnetV2'", ",", "create_aux_logits", "=", "True", ",",...
Creates the Inception Resnet V2 model. Args: inputs: a 4-D tensor of size [batch_size, height, width, 3]. nb_classes: number of predicted classes. is_training: whether is training or not. dropout_keep_prob: float, the fraction to keep before final layer. reuse: whether or not the network and its ...
[ "Creates", "the", "Inception", "Resnet", "V2", "model", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py#L288-L352
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py
inception_resnet_v2_arg_scope
def inception_resnet_v2_arg_scope(weight_decay=0.00004, batch_norm_decay=0.9997, batch_norm_epsilon=0.001): """Returns the scope with the default parameters for inception_resnet_v2. Args: weight_decay: the weight decay for weights variables. ...
python
def inception_resnet_v2_arg_scope(weight_decay=0.00004, batch_norm_decay=0.9997, batch_norm_epsilon=0.001): """Returns the scope with the default parameters for inception_resnet_v2. Args: weight_decay: the weight decay for weights variables. ...
[ "def", "inception_resnet_v2_arg_scope", "(", "weight_decay", "=", "0.00004", ",", "batch_norm_decay", "=", "0.9997", ",", "batch_norm_epsilon", "=", "0.001", ")", ":", "# Set weight_decay for weights in conv2d and fully_connected layers.", "with", "slim", ".", "arg_scope", ...
Returns the scope with the default parameters for inception_resnet_v2. Args: weight_decay: the weight decay for weights variables. batch_norm_decay: decay for the moving average of batch_norm momentums. batch_norm_epsilon: small float added to variance to avoid dividing by zero. Returns: a arg_sco...
[ "Returns", "the", "scope", "with", "the", "default", "parameters", "for", "inception_resnet_v2", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/dev_toolkit/sample_defenses/ens_adv_inception_resnet_v2/inception_resnet_v2.py#L358-L384
train
tensorflow/cleverhans
tutorials/future/tf2/mnist_tutorial.py
ld_mnist
def ld_mnist(): """Load training and test data.""" def convert_types(image, label): image = tf.cast(image, tf.float32) image /= 255 return image, label dataset, info = tfds.load('mnist', data_dir='gs://tfds-data/datasets', with_info=True, as_supervised=True) mnist_train...
python
def ld_mnist(): """Load training and test data.""" def convert_types(image, label): image = tf.cast(image, tf.float32) image /= 255 return image, label dataset, info = tfds.load('mnist', data_dir='gs://tfds-data/datasets', with_info=True, as_supervised=True) mnist_train...
[ "def", "ld_mnist", "(", ")", ":", "def", "convert_types", "(", "image", ",", "label", ")", ":", "image", "=", "tf", ".", "cast", "(", "image", ",", "tf", ".", "float32", ")", "image", "/=", "255", "return", "image", ",", "label", "dataset", ",", "i...
Load training and test data.
[ "Load", "training", "and", "test", "data", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/tutorials/future/tf2/mnist_tutorial.py#L29-L42
train
tensorflow/cleverhans
cleverhans_tutorials/mnist_tutorial_keras.py
mnist_tutorial
def mnist_tutorial(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_epochs=NB_EPOCHS, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, testing=False, label_smoothing=0.1): """ MNIST CleverHans tutorial :param train_start: index of first t...
python
def mnist_tutorial(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_epochs=NB_EPOCHS, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, testing=False, label_smoothing=0.1): """ MNIST CleverHans tutorial :param train_start: index of first t...
[ "def", "mnist_tutorial", "(", "train_start", "=", "0", ",", "train_end", "=", "60000", ",", "test_start", "=", "0", ",", "test_end", "=", "10000", ",", "nb_epochs", "=", "NB_EPOCHS", ",", "batch_size", "=", "BATCH_SIZE", ",", "learning_rate", "=", "LEARNING_...
MNIST CleverHans tutorial :param train_start: index of first training set example :param train_end: index of last training set example :param test_start: index of first test set example :param test_end: index of last test set example :param nb_epochs: number of epochs to train model :param batch_size: size ...
[ "MNIST", "CleverHans", "tutorial", ":", "param", "train_start", ":", "index", "of", "first", "training", "set", "example", ":", "param", "train_end", ":", "index", "of", "last", "training", "set", "example", ":", "param", "test_start", ":", "index", "of", "f...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans_tutorials/mnist_tutorial_keras.py#L30-L167
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
main
def main(args): """Validate all submissions and copy them into place""" random.seed() temp_dir = tempfile.mkdtemp() logging.info('Created temporary directory: %s', temp_dir) validator = SubmissionValidator( source_dir=args.source_dir, target_dir=args.target_dir, temp_dir=temp_dir, do_c...
python
def main(args): """Validate all submissions and copy them into place""" random.seed() temp_dir = tempfile.mkdtemp() logging.info('Created temporary directory: %s', temp_dir) validator = SubmissionValidator( source_dir=args.source_dir, target_dir=args.target_dir, temp_dir=temp_dir, do_c...
[ "def", "main", "(", "args", ")", ":", "random", ".", "seed", "(", ")", "temp_dir", "=", "tempfile", ".", "mkdtemp", "(", ")", "logging", ".", "info", "(", "'Created temporary directory: %s'", ",", "temp_dir", ")", "validator", "=", "SubmissionValidator", "("...
Validate all submissions and copy them into place
[ "Validate", "all", "submissions", "and", "copy", "them", "into", "place" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L229-L243
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
ValidationStats._update_stat
def _update_stat(self, submission_type, increase_success, increase_fail): """Common method to update submission statistics.""" stat = self.stats.get(submission_type, (0, 0)) stat = (stat[0] + increase_success, stat[1] + increase_fail) self.stats[submission_type] = stat
python
def _update_stat(self, submission_type, increase_success, increase_fail): """Common method to update submission statistics.""" stat = self.stats.get(submission_type, (0, 0)) stat = (stat[0] + increase_success, stat[1] + increase_fail) self.stats[submission_type] = stat
[ "def", "_update_stat", "(", "self", ",", "submission_type", ",", "increase_success", ",", "increase_fail", ")", ":", "stat", "=", "self", ".", "stats", ".", "get", "(", "submission_type", ",", "(", "0", ",", "0", ")", ")", "stat", "=", "(", "stat", "["...
Common method to update submission statistics.
[ "Common", "method", "to", "update", "submission", "statistics", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L64-L68
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
ValidationStats.log_stats
def log_stats(self): """Print statistics into log.""" logging.info('Validation statistics: ') for k, v in iteritems(self.stats): logging.info('%s - %d valid out of %d total submissions', k, v[0], v[0] + v[1])
python
def log_stats(self): """Print statistics into log.""" logging.info('Validation statistics: ') for k, v in iteritems(self.stats): logging.info('%s - %d valid out of %d total submissions', k, v[0], v[0] + v[1])
[ "def", "log_stats", "(", "self", ")", ":", "logging", ".", "info", "(", "'Validation statistics: '", ")", "for", "k", ",", "v", "in", "iteritems", "(", "self", ".", "stats", ")", ":", "logging", ".", "info", "(", "'%s - %d valid out of %d total submissions'", ...
Print statistics into log.
[ "Print", "statistics", "into", "log", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L78-L83
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
SubmissionValidator.copy_submission_locally
def copy_submission_locally(self, cloud_path): """Copies submission from Google Cloud Storage to local directory. Args: cloud_path: path of the submission in Google Cloud Storage Returns: name of the local file where submission is copied to """ local_path = os.path.join(self.download_d...
python
def copy_submission_locally(self, cloud_path): """Copies submission from Google Cloud Storage to local directory. Args: cloud_path: path of the submission in Google Cloud Storage Returns: name of the local file where submission is copied to """ local_path = os.path.join(self.download_d...
[ "def", "copy_submission_locally", "(", "self", ",", "cloud_path", ")", ":", "local_path", "=", "os", ".", "path", ".", "join", "(", "self", ".", "download_dir", ",", "os", ".", "path", ".", "basename", "(", "cloud_path", ")", ")", "cmd", "=", "[", "'gs...
Copies submission from Google Cloud Storage to local directory. Args: cloud_path: path of the submission in Google Cloud Storage Returns: name of the local file where submission is copied to
[ "Copies", "submission", "from", "Google", "Cloud", "Storage", "to", "local", "directory", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L119-L133
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
SubmissionValidator.copy_submission_to_destination
def copy_submission_to_destination(self, src_filename, dst_subdir, submission_id): """Copies submission to target directory. Args: src_filename: source filename of the submission dst_subdir: subdirectory of the target directory where submission should be...
python
def copy_submission_to_destination(self, src_filename, dst_subdir, submission_id): """Copies submission to target directory. Args: src_filename: source filename of the submission dst_subdir: subdirectory of the target directory where submission should be...
[ "def", "copy_submission_to_destination", "(", "self", ",", "src_filename", ",", "dst_subdir", ",", "submission_id", ")", ":", "extension", "=", "[", "e", "for", "e", "in", "ALLOWED_EXTENSIONS", "if", "src_filename", ".", "endswith", "(", "e", ")", "]", "if", ...
Copies submission to target directory. Args: src_filename: source filename of the submission dst_subdir: subdirectory of the target directory where submission should be copied to submission_id: ID of the submission, will be used as a new submission filename (before extension)
[ "Copies", "submission", "to", "target", "directory", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L135-L157
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
SubmissionValidator.validate_and_copy_one_submission
def validate_and_copy_one_submission(self, submission_path): """Validates one submission and copies it to target directory. Args: submission_path: path in Google Cloud Storage of the submission file """ if os.path.exists(self.download_dir): shutil.rmtree(self.download_dir) os.makedirs(s...
python
def validate_and_copy_one_submission(self, submission_path): """Validates one submission and copies it to target directory. Args: submission_path: path in Google Cloud Storage of the submission file """ if os.path.exists(self.download_dir): shutil.rmtree(self.download_dir) os.makedirs(s...
[ "def", "validate_and_copy_one_submission", "(", "self", ",", "submission_path", ")", ":", "if", "os", ".", "path", ".", "exists", "(", "self", ".", "download_dir", ")", ":", "shutil", ".", "rmtree", "(", "self", ".", "download_dir", ")", "os", ".", "makedi...
Validates one submission and copies it to target directory. Args: submission_path: path in Google Cloud Storage of the submission file
[ "Validates", "one", "submission", "and", "copies", "it", "to", "target", "directory", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L159-L190
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
SubmissionValidator.save_id_to_path_mapping
def save_id_to_path_mapping(self): """Saves mapping from submission IDs to original filenames. This mapping is saved as CSV file into target directory. """ if not self.id_to_path_mapping: return with open(self.local_id_to_path_mapping_file, 'w') as f: writer = csv.writer(f) writer...
python
def save_id_to_path_mapping(self): """Saves mapping from submission IDs to original filenames. This mapping is saved as CSV file into target directory. """ if not self.id_to_path_mapping: return with open(self.local_id_to_path_mapping_file, 'w') as f: writer = csv.writer(f) writer...
[ "def", "save_id_to_path_mapping", "(", "self", ")", ":", "if", "not", "self", ".", "id_to_path_mapping", ":", "return", "with", "open", "(", "self", ".", "local_id_to_path_mapping_file", ",", "'w'", ")", "as", "f", ":", "writer", "=", "csv", ".", "writer", ...
Saves mapping from submission IDs to original filenames. This mapping is saved as CSV file into target directory.
[ "Saves", "mapping", "from", "submission", "IDs", "to", "original", "filenames", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L192-L207
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py
SubmissionValidator.run
def run(self): """Runs validation of all submissions.""" cmd = ['gsutil', 'ls', os.path.join(self.source_dir, '**')] try: files_list = subprocess.check_output(cmd).split('\n') except subprocess.CalledProcessError: logging.error('Can''t read source directory') all_submissions = [ ...
python
def run(self): """Runs validation of all submissions.""" cmd = ['gsutil', 'ls', os.path.join(self.source_dir, '**')] try: files_list = subprocess.check_output(cmd).split('\n') except subprocess.CalledProcessError: logging.error('Can''t read source directory') all_submissions = [ ...
[ "def", "run", "(", "self", ")", ":", "cmd", "=", "[", "'gsutil'", ",", "'ls'", ",", "os", ".", "path", ".", "join", "(", "self", ".", "source_dir", ",", "'**'", ")", "]", "try", ":", "files_list", "=", "subprocess", ".", "check_output", "(", "cmd",...
Runs validation of all submissions.
[ "Runs", "validation", "of", "all", "submissions", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/validation_tool/validate_and_copy_submissions.py#L209-L226
train
tensorflow/cleverhans
scripts/plot_success_fail_curve.py
main
def main(argv=None): """Takes the path to a directory with reports and renders success fail plots.""" report_paths = argv[1:] fail_names = FLAGS.fail_names.split(',') for report_path in report_paths: plot_report_from_path(report_path, label=report_path, fail_names=fail_names) pyplot.legend() pyplot.x...
python
def main(argv=None): """Takes the path to a directory with reports and renders success fail plots.""" report_paths = argv[1:] fail_names = FLAGS.fail_names.split(',') for report_path in report_paths: plot_report_from_path(report_path, label=report_path, fail_names=fail_names) pyplot.legend() pyplot.x...
[ "def", "main", "(", "argv", "=", "None", ")", ":", "report_paths", "=", "argv", "[", "1", ":", "]", "fail_names", "=", "FLAGS", ".", "fail_names", ".", "split", "(", "','", ")", "for", "report_path", "in", "report_paths", ":", "plot_report_from_path", "(...
Takes the path to a directory with reports and renders success fail plots.
[ "Takes", "the", "path", "to", "a", "directory", "with", "reports", "and", "renders", "success", "fail", "plots", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/scripts/plot_success_fail_curve.py#L25-L38
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
is_unclaimed
def is_unclaimed(work): """Returns True if work piece is unclaimed.""" if work['is_completed']: return False cutoff_time = time.time() - MAX_PROCESSING_TIME if (work['claimed_worker_id'] and work['claimed_worker_start_time'] is not None and work['claimed_worker_start_time'] >= cutoff_time): ...
python
def is_unclaimed(work): """Returns True if work piece is unclaimed.""" if work['is_completed']: return False cutoff_time = time.time() - MAX_PROCESSING_TIME if (work['claimed_worker_id'] and work['claimed_worker_start_time'] is not None and work['claimed_worker_start_time'] >= cutoff_time): ...
[ "def", "is_unclaimed", "(", "work", ")", ":", "if", "work", "[", "'is_completed'", "]", ":", "return", "False", "cutoff_time", "=", "time", ".", "time", "(", ")", "-", "MAX_PROCESSING_TIME", "if", "(", "work", "[", "'claimed_worker_id'", "]", "and", "work"...
Returns True if work piece is unclaimed.
[ "Returns", "True", "if", "work", "piece", "is", "unclaimed", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L46-L55
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
WorkPiecesBase.write_all_to_datastore
def write_all_to_datastore(self): """Writes all work pieces into datastore. Each work piece is identified by ID. This method writes/updates only those work pieces which IDs are stored in this class. For examples, if this class has only work pieces with IDs '1' ... '100' and datastore already contains ...
python
def write_all_to_datastore(self): """Writes all work pieces into datastore. Each work piece is identified by ID. This method writes/updates only those work pieces which IDs are stored in this class. For examples, if this class has only work pieces with IDs '1' ... '100' and datastore already contains ...
[ "def", "write_all_to_datastore", "(", "self", ")", ":", "client", "=", "self", ".", "_datastore_client", "with", "client", ".", "no_transact_batch", "(", ")", "as", "batch", ":", "parent_key", "=", "client", ".", "key", "(", "KIND_WORK_TYPE", ",", "self", "....
Writes all work pieces into datastore. Each work piece is identified by ID. This method writes/updates only those work pieces which IDs are stored in this class. For examples, if this class has only work pieces with IDs '1' ... '100' and datastore already contains work pieces with IDs '50' ... '200' t...
[ "Writes", "all", "work", "pieces", "into", "datastore", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L150-L168
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
WorkPiecesBase.read_all_from_datastore
def read_all_from_datastore(self): """Reads all work pieces from the datastore.""" self._work = {} client = self._datastore_client parent_key = client.key(KIND_WORK_TYPE, self._work_type_entity_id) for entity in client.query_fetch(kind=KIND_WORK, ancestor=parent_key): work_id = entity.key.flat...
python
def read_all_from_datastore(self): """Reads all work pieces from the datastore.""" self._work = {} client = self._datastore_client parent_key = client.key(KIND_WORK_TYPE, self._work_type_entity_id) for entity in client.query_fetch(kind=KIND_WORK, ancestor=parent_key): work_id = entity.key.flat...
[ "def", "read_all_from_datastore", "(", "self", ")", ":", "self", ".", "_work", "=", "{", "}", "client", "=", "self", ".", "_datastore_client", "parent_key", "=", "client", ".", "key", "(", "KIND_WORK_TYPE", ",", "self", ".", "_work_type_entity_id", ")", "for...
Reads all work pieces from the datastore.
[ "Reads", "all", "work", "pieces", "from", "the", "datastore", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L170-L177
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
WorkPiecesBase._read_undone_shard_from_datastore
def _read_undone_shard_from_datastore(self, shard_id=None): """Reads undone worke pieces which are assigned to shard with given id.""" self._work = {} client = self._datastore_client parent_key = client.key(KIND_WORK_TYPE, self._work_type_entity_id) filters = [('is_completed', '=', False)] if sh...
python
def _read_undone_shard_from_datastore(self, shard_id=None): """Reads undone worke pieces which are assigned to shard with given id.""" self._work = {} client = self._datastore_client parent_key = client.key(KIND_WORK_TYPE, self._work_type_entity_id) filters = [('is_completed', '=', False)] if sh...
[ "def", "_read_undone_shard_from_datastore", "(", "self", ",", "shard_id", "=", "None", ")", ":", "self", ".", "_work", "=", "{", "}", "client", "=", "self", ".", "_datastore_client", "parent_key", "=", "client", ".", "key", "(", "KIND_WORK_TYPE", ",", "self"...
Reads undone worke pieces which are assigned to shard with given id.
[ "Reads", "undone", "worke", "pieces", "which", "are", "assigned", "to", "shard", "with", "given", "id", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L179-L192
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
WorkPiecesBase.read_undone_from_datastore
def read_undone_from_datastore(self, shard_id=None, num_shards=None): """Reads undone work from the datastore. If shard_id and num_shards are specified then this method will attempt to read undone work for shard with id shard_id. If no undone work was found then it will try to read shard (shard_id+1) a...
python
def read_undone_from_datastore(self, shard_id=None, num_shards=None): """Reads undone work from the datastore. If shard_id and num_shards are specified then this method will attempt to read undone work for shard with id shard_id. If no undone work was found then it will try to read shard (shard_id+1) a...
[ "def", "read_undone_from_datastore", "(", "self", ",", "shard_id", "=", "None", ",", "num_shards", "=", "None", ")", ":", "if", "shard_id", "is", "not", "None", ":", "shards_list", "=", "[", "(", "i", "+", "shard_id", ")", "%", "num_shards", "for", "i", ...
Reads undone work from the datastore. If shard_id and num_shards are specified then this method will attempt to read undone work for shard with id shard_id. If no undone work was found then it will try to read shard (shard_id+1) and so on until either found shard with undone work or all shards are read...
[ "Reads", "undone", "work", "from", "the", "datastore", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L194-L219
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
WorkPiecesBase.try_pick_piece_of_work
def try_pick_piece_of_work(self, worker_id, submission_id=None): """Tries pick next unclaimed piece of work to do. Attempt to claim work piece is done using Cloud Datastore transaction, so only one worker can claim any work piece at a time. Args: worker_id: ID of current worker submission_...
python
def try_pick_piece_of_work(self, worker_id, submission_id=None): """Tries pick next unclaimed piece of work to do. Attempt to claim work piece is done using Cloud Datastore transaction, so only one worker can claim any work piece at a time. Args: worker_id: ID of current worker submission_...
[ "def", "try_pick_piece_of_work", "(", "self", ",", "worker_id", ",", "submission_id", "=", "None", ")", ":", "client", "=", "self", ".", "_datastore_client", "unclaimed_work_ids", "=", "None", "if", "submission_id", ":", "unclaimed_work_ids", "=", "[", "k", "for...
Tries pick next unclaimed piece of work to do. Attempt to claim work piece is done using Cloud Datastore transaction, so only one worker can claim any work piece at a time. Args: worker_id: ID of current worker submission_id: if not None then this method will try to pick piece of work ...
[ "Tries", "pick", "next", "unclaimed", "piece", "of", "work", "to", "do", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L221-L261
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
WorkPiecesBase.update_work_as_completed
def update_work_as_completed(self, worker_id, work_id, other_values=None, error=None): """Updates work piece in datastore as completed. Args: worker_id: ID of the worker which did the work work_id: ID of the work which was done other_values: dictionary with addi...
python
def update_work_as_completed(self, worker_id, work_id, other_values=None, error=None): """Updates work piece in datastore as completed. Args: worker_id: ID of the worker which did the work work_id: ID of the work which was done other_values: dictionary with addi...
[ "def", "update_work_as_completed", "(", "self", ",", "worker_id", ",", "work_id", ",", "other_values", "=", "None", ",", "error", "=", "None", ")", ":", "client", "=", "self", ".", "_datastore_client", "try", ":", "with", "client", ".", "transaction", "(", ...
Updates work piece in datastore as completed. Args: worker_id: ID of the worker which did the work work_id: ID of the work which was done other_values: dictionary with additonal values which should be saved with the work piece error: if not None then error occurred during computatio...
[ "Updates", "work", "piece", "in", "datastore", "as", "completed", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L263-L294
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
WorkPiecesBase.compute_work_statistics
def compute_work_statistics(self): """Computes statistics from all work pieces stored in this class.""" result = {} for v in itervalues(self.work): submission_id = v['submission_id'] if submission_id not in result: result[submission_id] = { 'completed': 0, 'num_er...
python
def compute_work_statistics(self): """Computes statistics from all work pieces stored in this class.""" result = {} for v in itervalues(self.work): submission_id = v['submission_id'] if submission_id not in result: result[submission_id] = { 'completed': 0, 'num_er...
[ "def", "compute_work_statistics", "(", "self", ")", ":", "result", "=", "{", "}", "for", "v", "in", "itervalues", "(", "self", ".", "work", ")", ":", "submission_id", "=", "v", "[", "'submission_id'", "]", "if", "submission_id", "not", "in", "result", ":...
Computes statistics from all work pieces stored in this class.
[ "Computes", "statistics", "from", "all", "work", "pieces", "stored", "in", "this", "class", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L296-L326
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
AttackWorkPieces.init_from_adversarial_batches
def init_from_adversarial_batches(self, adv_batches): """Initializes work pieces from adversarial batches. Args: adv_batches: dict with adversarial batches, could be obtained as AversarialBatches.data """ for idx, (adv_batch_id, adv_batch_val) in enumerate(iteritems(adv_batches)): w...
python
def init_from_adversarial_batches(self, adv_batches): """Initializes work pieces from adversarial batches. Args: adv_batches: dict with adversarial batches, could be obtained as AversarialBatches.data """ for idx, (adv_batch_id, adv_batch_val) in enumerate(iteritems(adv_batches)): w...
[ "def", "init_from_adversarial_batches", "(", "self", ",", "adv_batches", ")", ":", "for", "idx", ",", "(", "adv_batch_id", ",", "adv_batch_val", ")", "in", "enumerate", "(", "iteritems", "(", "adv_batches", ")", ")", ":", "work_id", "=", "ATTACK_WORK_ID_PATTERN"...
Initializes work pieces from adversarial batches. Args: adv_batches: dict with adversarial batches, could be obtained as AversarialBatches.data
[ "Initializes", "work", "pieces", "from", "adversarial", "batches", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L349-L367
train
tensorflow/cleverhans
examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py
DefenseWorkPieces.init_from_class_batches
def init_from_class_batches(self, class_batches, num_shards=None): """Initializes work pieces from classification batches. Args: class_batches: dict with classification batches, could be obtained as ClassificationBatches.data num_shards: number of shards to split data into, if None ...
python
def init_from_class_batches(self, class_batches, num_shards=None): """Initializes work pieces from classification batches. Args: class_batches: dict with classification batches, could be obtained as ClassificationBatches.data num_shards: number of shards to split data into, if None ...
[ "def", "init_from_class_batches", "(", "self", ",", "class_batches", ",", "num_shards", "=", "None", ")", ":", "shards_for_submissions", "=", "{", "}", "shard_idx", "=", "0", "for", "idx", ",", "(", "batch_id", ",", "batch_val", ")", "in", "enumerate", "(", ...
Initializes work pieces from classification batches. Args: class_batches: dict with classification batches, could be obtained as ClassificationBatches.data num_shards: number of shards to split data into, if None then no sharding is done.
[ "Initializes", "work", "pieces", "from", "classification", "batches", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/examples/nips17_adversarial_competition/eval_infra/code/eval_lib/work_data.py#L379-L411
train
tensorflow/cleverhans
scripts/make_confidence_report_bundle_examples.py
main
def main(argv=None): """ Make a confidence report and save it to disk. """ assert len(argv) >= 3 _name_of_script = argv[0] model_filepath = argv[1] adv_x_filepaths = argv[2:] sess = tf.Session() with sess.as_default(): model = serial.load(model_filepath) factory = model.dataset_factory facto...
python
def main(argv=None): """ Make a confidence report and save it to disk. """ assert len(argv) >= 3 _name_of_script = argv[0] model_filepath = argv[1] adv_x_filepaths = argv[2:] sess = tf.Session() with sess.as_default(): model = serial.load(model_filepath) factory = model.dataset_factory facto...
[ "def", "main", "(", "argv", "=", "None", ")", ":", "assert", "len", "(", "argv", ")", ">=", "3", "_name_of_script", "=", "argv", "[", "0", "]", "model_filepath", "=", "argv", "[", "1", "]", "adv_x_filepaths", "=", "argv", "[", "2", ":", "]", "sess"...
Make a confidence report and save it to disk.
[ "Make", "a", "confidence", "report", "and", "save", "it", "to", "disk", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/scripts/make_confidence_report_bundle_examples.py#L48-L95
train
tensorflow/cleverhans
cleverhans/attacks/fast_gradient_method.py
fgm
def fgm(x, logits, y=None, eps=0.3, ord=np.inf, clip_min=None, clip_max=None, targeted=False, sanity_checks=True): """ TensorFlow implementation of the Fast Gradient Method. :param x: the input placeholder :param logits: output of model.get_logits ...
python
def fgm(x, logits, y=None, eps=0.3, ord=np.inf, clip_min=None, clip_max=None, targeted=False, sanity_checks=True): """ TensorFlow implementation of the Fast Gradient Method. :param x: the input placeholder :param logits: output of model.get_logits ...
[ "def", "fgm", "(", "x", ",", "logits", ",", "y", "=", "None", ",", "eps", "=", "0.3", ",", "ord", "=", "np", ".", "inf", ",", "clip_min", "=", "None", ",", "clip_max", "=", "None", ",", "targeted", "=", "False", ",", "sanity_checks", "=", "True",...
TensorFlow implementation of the Fast Gradient Method. :param x: the input placeholder :param logits: output of model.get_logits :param y: (optional) A placeholder for the true labels. If targeted is true, then provide the target label. Otherwise, only provide this parameter if you'd like ...
[ "TensorFlow", "implementation", "of", "the", "Fast", "Gradient", "Method", ".", ":", "param", "x", ":", "the", "input", "placeholder", ":", "param", "logits", ":", "output", "of", "model", ".", "get_logits", ":", "param", "y", ":", "(", "optional", ")", ...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/fast_gradient_method.py#L119-L194
train
tensorflow/cleverhans
cleverhans/attacks/fast_gradient_method.py
optimize_linear
def optimize_linear(grad, eps, ord=np.inf): """ Solves for the optimal input to a linear function under a norm constraint. Optimal_perturbation = argmax_{eta, ||eta||_{ord} < eps} dot(eta, grad) :param grad: tf tensor containing a batch of gradients :param eps: float scalar specifying size of constraint reg...
python
def optimize_linear(grad, eps, ord=np.inf): """ Solves for the optimal input to a linear function under a norm constraint. Optimal_perturbation = argmax_{eta, ||eta||_{ord} < eps} dot(eta, grad) :param grad: tf tensor containing a batch of gradients :param eps: float scalar specifying size of constraint reg...
[ "def", "optimize_linear", "(", "grad", ",", "eps", ",", "ord", "=", "np", ".", "inf", ")", ":", "# In Python 2, the `list` call in the following line is redundant / harmless.", "# In Python 3, the `list` call is needed to convert the iterator returned by `range` into a list.", "red_i...
Solves for the optimal input to a linear function under a norm constraint. Optimal_perturbation = argmax_{eta, ||eta||_{ord} < eps} dot(eta, grad) :param grad: tf tensor containing a batch of gradients :param eps: float scalar specifying size of constraint region :param ord: int specifying order of norm :re...
[ "Solves", "for", "the", "optimal", "input", "to", "a", "linear", "function", "under", "a", "norm", "constraint", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/fast_gradient_method.py#L197-L243
train
tensorflow/cleverhans
cleverhans/attacks/fast_gradient_method.py
FastGradientMethod.generate
def generate(self, x, **kwargs): """ Returns the graph for Fast Gradient Method adversarial examples. :param x: The model's symbolic inputs. :param kwargs: See `parse_params` """ # Parse and save attack-specific parameters assert self.parse_params(**kwargs) labels, _nb_classes = self.g...
python
def generate(self, x, **kwargs): """ Returns the graph for Fast Gradient Method adversarial examples. :param x: The model's symbolic inputs. :param kwargs: See `parse_params` """ # Parse and save attack-specific parameters assert self.parse_params(**kwargs) labels, _nb_classes = self.g...
[ "def", "generate", "(", "self", ",", "x", ",", "*", "*", "kwargs", ")", ":", "# Parse and save attack-specific parameters", "assert", "self", ".", "parse_params", "(", "*", "*", "kwargs", ")", "labels", ",", "_nb_classes", "=", "self", ".", "get_or_guess_label...
Returns the graph for Fast Gradient Method adversarial examples. :param x: The model's symbolic inputs. :param kwargs: See `parse_params`
[ "Returns", "the", "graph", "for", "Fast", "Gradient", "Method", "adversarial", "examples", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/fast_gradient_method.py#L40-L61
train
tensorflow/cleverhans
cleverhans/experimental/certification/nn.py
load_network_from_checkpoint
def load_network_from_checkpoint(checkpoint, model_json, input_shape=None): """Function to read the weights from checkpoint based on json description. Args: checkpoint: tensorflow checkpoint with trained model to verify model_json: path of json file with model description of the netwo...
python
def load_network_from_checkpoint(checkpoint, model_json, input_shape=None): """Function to read the weights from checkpoint based on json description. Args: checkpoint: tensorflow checkpoint with trained model to verify model_json: path of json file with model description of the netwo...
[ "def", "load_network_from_checkpoint", "(", "checkpoint", ",", "model_json", ",", "input_shape", "=", "None", ")", ":", "# Load checkpoint", "reader", "=", "tf", ".", "train", ".", "load_checkpoint", "(", "checkpoint", ")", "variable_map", "=", "reader", ".", "g...
Function to read the weights from checkpoint based on json description. Args: checkpoint: tensorflow checkpoint with trained model to verify model_json: path of json file with model description of the network list of dictionary items for each layer containing 'type', 'weight_var...
[ "Function", "to", "read", "the", "weights", "from", "checkpoint", "based", "on", "json", "description", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/experimental/certification/nn.py#L161-L226
train
tensorflow/cleverhans
cleverhans/experimental/certification/nn.py
NeuralNetwork.forward_pass
def forward_pass(self, vector, layer_index, is_transpose=False, is_abs=False): """Performs forward pass through the layer weights at layer_index. Args: vector: vector that has to be passed through in forward pass layer_index: index of the layer is_transpose: whether the weights of the layer h...
python
def forward_pass(self, vector, layer_index, is_transpose=False, is_abs=False): """Performs forward pass through the layer weights at layer_index. Args: vector: vector that has to be passed through in forward pass layer_index: index of the layer is_transpose: whether the weights of the layer h...
[ "def", "forward_pass", "(", "self", ",", "vector", ",", "layer_index", ",", "is_transpose", "=", "False", ",", "is_abs", "=", "False", ")", ":", "if", "(", "layer_index", "<", "0", "or", "layer_index", ">", "self", ".", "num_hidden_layers", ")", ":", "ra...
Performs forward pass through the layer weights at layer_index. Args: vector: vector that has to be passed through in forward pass layer_index: index of the layer is_transpose: whether the weights of the layer have to be transposed is_abs: whether to take the absolute value of the weights ...
[ "Performs", "forward", "pass", "through", "the", "layer", "weights", "at", "layer_index", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/experimental/certification/nn.py#L111-L159
train
tensorflow/cleverhans
cleverhans/devtools/version.py
dev_version
def dev_version(): """ Returns a hexdigest of all the python files in the module. """ md5_hash = hashlib.md5() py_files = sorted(list_files(suffix=".py")) if not py_files: return '' for filename in py_files: with open(filename, 'rb') as fobj: content = fobj.read() md5_hash.update(conten...
python
def dev_version(): """ Returns a hexdigest of all the python files in the module. """ md5_hash = hashlib.md5() py_files = sorted(list_files(suffix=".py")) if not py_files: return '' for filename in py_files: with open(filename, 'rb') as fobj: content = fobj.read() md5_hash.update(conten...
[ "def", "dev_version", "(", ")", ":", "md5_hash", "=", "hashlib", ".", "md5", "(", ")", "py_files", "=", "sorted", "(", "list_files", "(", "suffix", "=", "\".py\"", ")", ")", "if", "not", "py_files", ":", "return", "''", "for", "filename", "in", "py_fil...
Returns a hexdigest of all the python files in the module.
[ "Returns", "a", "hexdigest", "of", "all", "the", "python", "files", "in", "the", "module", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/devtools/version.py#L11-L24
train
tensorflow/cleverhans
scripts/print_report.py
current
def current(report): """ The current implementation of report printing. :param report: ConfidenceReport """ if hasattr(report, "completed"): if report.completed: print("Report completed") else: print("REPORT NOT COMPLETED") else: warnings.warn("This report does not indicate whether i...
python
def current(report): """ The current implementation of report printing. :param report: ConfidenceReport """ if hasattr(report, "completed"): if report.completed: print("Report completed") else: print("REPORT NOT COMPLETED") else: warnings.warn("This report does not indicate whether i...
[ "def", "current", "(", "report", ")", ":", "if", "hasattr", "(", "report", ",", "\"completed\"", ")", ":", "if", "report", ".", "completed", ":", "print", "(", "\"Report completed\"", ")", "else", ":", "print", "(", "\"REPORT NOT COMPLETED\"", ")", "else", ...
The current implementation of report printing. :param report: ConfidenceReport
[ "The", "current", "implementation", "of", "report", "printing", ".", ":", "param", "report", ":", "ConfidenceReport" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/scripts/print_report.py#L23-L41
train
tensorflow/cleverhans
scripts/print_report.py
deprecated
def deprecated(report): """ The deprecated implementation of report printing. :param report: dict """ warnings.warn("Printing dict-based reports is deprecated. This function " "is included only to support a private development branch " "and may be removed without warning.") ...
python
def deprecated(report): """ The deprecated implementation of report printing. :param report: dict """ warnings.warn("Printing dict-based reports is deprecated. This function " "is included only to support a private development branch " "and may be removed without warning.") ...
[ "def", "deprecated", "(", "report", ")", ":", "warnings", ".", "warn", "(", "\"Printing dict-based reports is deprecated. This function \"", "\"is included only to support a private development branch \"", "\"and may be removed without warning.\"", ")", "for", "key", "in", "report"...
The deprecated implementation of report printing. :param report: dict
[ "The", "deprecated", "implementation", "of", "report", "printing", ".", ":", "param", "report", ":", "dict" ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/scripts/print_report.py#L43-L65
train
tensorflow/cleverhans
cleverhans/model_zoo/soft_nearest_neighbor_loss/SNNL_regularized_train.py
SNNL_example
def SNNL_example(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_epochs=NB_EPOCHS, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, nb_filters=NB_FILTERS, SNNL_factor=SNNL_FACTOR, output_dir=OUTPUT_DIR): """ A s...
python
def SNNL_example(train_start=0, train_end=60000, test_start=0, test_end=10000, nb_epochs=NB_EPOCHS, batch_size=BATCH_SIZE, learning_rate=LEARNING_RATE, nb_filters=NB_FILTERS, SNNL_factor=SNNL_FACTOR, output_dir=OUTPUT_DIR): """ A s...
[ "def", "SNNL_example", "(", "train_start", "=", "0", ",", "train_end", "=", "60000", ",", "test_start", "=", "0", ",", "test_end", "=", "10000", ",", "nb_epochs", "=", "NB_EPOCHS", ",", "batch_size", "=", "BATCH_SIZE", ",", "learning_rate", "=", "LEARNING_RA...
A simple model trained to minimize Cross Entropy and Maximize Soft Nearest Neighbor Loss at each internal layer. This outputs a TSNE of the sign of the adversarial gradients of a trained model. A model with a negative SNNL_factor will show little or no class clusters, while a model with a 0 SNNL_factor will hav...
[ "A", "simple", "model", "trained", "to", "minimize", "Cross", "Entropy", "and", "Maximize", "Soft", "Nearest", "Neighbor", "Loss", "at", "each", "internal", "layer", ".", "This", "outputs", "a", "TSNE", "of", "the", "sign", "of", "the", "adversarial", "gradi...
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/model_zoo/soft_nearest_neighbor_loss/SNNL_regularized_train.py#L37-L149
train
tensorflow/cleverhans
cleverhans/attacks/projected_gradient_descent.py
ProjectedGradientDescent.generate
def generate(self, x, **kwargs): """ Generate symbolic graph for adversarial examples and return. :param x: The model's symbolic inputs. :param kwargs: See `parse_params` """ # Parse and save attack-specific parameters assert self.parse_params(**kwargs) asserts = [] # If a data ra...
python
def generate(self, x, **kwargs): """ Generate symbolic graph for adversarial examples and return. :param x: The model's symbolic inputs. :param kwargs: See `parse_params` """ # Parse and save attack-specific parameters assert self.parse_params(**kwargs) asserts = [] # If a data ra...
[ "def", "generate", "(", "self", ",", "x", ",", "*", "*", "kwargs", ")", ":", "# Parse and save attack-specific parameters", "assert", "self", ".", "parse_params", "(", "*", "*", "kwargs", ")", "asserts", "=", "[", "]", "# If a data range was specified, check that ...
Generate symbolic graph for adversarial examples and return. :param x: The model's symbolic inputs. :param kwargs: See `parse_params`
[ "Generate", "symbolic", "graph", "for", "adversarial", "examples", "and", "return", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/projected_gradient_descent.py#L48-L166
train
tensorflow/cleverhans
cleverhans/attacks/projected_gradient_descent.py
ProjectedGradientDescent.parse_params
def parse_params(self, eps=0.3, eps_iter=0.05, nb_iter=10, y=None, ord=np.inf, clip_min=None, clip_max=None, y_target=None, rand_init=None, ...
python
def parse_params(self, eps=0.3, eps_iter=0.05, nb_iter=10, y=None, ord=np.inf, clip_min=None, clip_max=None, y_target=None, rand_init=None, ...
[ "def", "parse_params", "(", "self", ",", "eps", "=", "0.3", ",", "eps_iter", "=", "0.05", ",", "nb_iter", "=", "10", ",", "y", "=", "None", ",", "ord", "=", "np", ".", "inf", ",", "clip_min", "=", "None", ",", "clip_max", "=", "None", ",", "y_tar...
Take in a dictionary of parameters and applies attack-specific checks before saving them as attributes. Attack-specific parameters: :param eps: (optional float) maximum distortion of adversarial example compared to original input :param eps_iter: (optional float) step size for each att...
[ "Take", "in", "a", "dictionary", "of", "parameters", "and", "applies", "attack", "-", "specific", "checks", "before", "saving", "them", "as", "attributes", "." ]
97488e215760547b81afc53f5e5de8ba7da5bd98
https://github.com/tensorflow/cleverhans/blob/97488e215760547b81afc53f5e5de8ba7da5bd98/cleverhans/attacks/projected_gradient_descent.py#L168-L237
train