fgsm_attack = FastGradientMethod(wrap, sess=sess)
X_adv = np.zeros((len(indices), 32, 32, 3))
for i in range(0, len(indices), 200):
X_adv[i:(i + 200)] = fgsm_attack.generate_np(X_test[indices[i:(i + 200)]],
**fgsm_params)
print("metrics")
print(metrics(model, X_adv, X_test, y_test, indices))
print(metrics(model_target, X_adv, X_test, y_test, indices))
#MIM Diverse
print("MIM-DIVERSE")
mim_params = {
'eps': 0.03,
'eps_iter': 0.01,
'nb_iter': 40,
'ord': np.inf,
'clip_min': 0.,
'clip_max': 1.,
'prob': 0.8
}
mim_attack = MomentumIterativeMethod_Diverse(wrap, sess=sess)
X_adv = np.zeros((len(indices), 32, 32, 3))
for i in np.arange(0, len(indices), 200):
X_adv[i:(i + 200)] = mim_attack.generate_np(X_test[indices[i:(i + 200)]],
**mim_params)
print("metrics")
print(metrics(model, X_adv, X_test, y_test, indices))
print(metrics(model_target, X_adv, X_test, y_test, indices))
print("MIMDiverse")
mim_params = {
'eps': float(sys.argv[1]),
'eps_iter': 0.01,
'nb_iter': 500,
'ord': np.inf,
'clip_min': 0.,
'clip_max': 1.
}
for i in np.arange(0.1, 1.1, 0.1):
i = np.float32(i)
mim_params['prob'] = i
print(i)
mim_source = MomentumIterativeMethod_Diverse(wrap_source, sess=sess)
X_adv_source = np.zeros((len(indices_test), 32, 32, 3))
for i in np.arange(0, len(indices_test), 500):
X_adv_source[i:(i + 500)] = mim_source.generate_np(
X_test[indices_test[i:(i + 500)]], **mim_params)
print("metrics source model")
print(
metrics(model_source, X_adv_source, X_test, pred_source, indices_test))
print("metrics base model")
print(metrics(model, X_adv_source, X_test, pred_base, indices_test))
pred_source_adv = np.argmax(model_source.predict(X_adv_source), axis=1)
pred_adv_basefromsource = np.argmax(model.predict(X_adv_source), axis=1)
agree_func(indices_test, pred_adv_basefromsource, pred_source_adv,