Nataniel Ruiz
50 lines
1.3 KiB
Python
50 lines
1.3 KiB
Python
import copy
|
|
import numpy as np
|
|
from collections import Iterable
|
|
from scipy.stats import truncnorm
|
|
|
|
import torch
|
|
import torch.nn as nn
|
|
|
|
class LinfPGDAttack(object):
|
|
def __init__(self, model=None, epsilon=0.05, k=1, a=0.05):
|
|
self.model = model
|
|
self.epsilon = epsilon
|
|
self.k = k
|
|
self.a = a
|
|
self.loss_fn = nn.MSELoss()
|
|
|
|
def perturb(self, X_nat, y):
|
|
"""
|
|
Given examples (X_nat, y), returns adversarial
|
|
examples within epsilon of X_nat in l_infinity norm.
|
|
"""
|
|
X = X_nat.clone().detach_()
|
|
|
|
for i in range(self.k):
|
|
print('test', i)
|
|
X.requires_grad = True
|
|
output = self.model(X)
|
|
|
|
self.model.zero_grad()
|
|
loss = -self.loss_fn(output, y)
|
|
loss.backward()
|
|
grad = X.grad
|
|
|
|
X_adv = X + self.a * grad.sign()
|
|
|
|
eta = torch.clamp(X_adv - X_nat, min=-self.epsilon, max=self.epsilon)
|
|
X = torch.clamp(X_nat + eta, min=-1, max=1).detach_()
|
|
eta = None
|
|
X_adv = None
|
|
|
|
return X, X - X_nat
|
|
|
|
def clip_tensor(X, Y, Z):
|
|
# Clip X with Y min and Z max
|
|
X_np = X.data.cpu().numpy()
|
|
Y_np = Y.data.cpu().numpy()
|
|
Z_np = Z.data.cpu().numpy()
|
|
X_clipped = np.clip(X_np, Y_np, Z_np)
|
|
X_res = torch.FloatTensor(X_clipped)
|
|
return X_res |