Questions tagged [adversarial-attack]
17 questions
1
vote
0 answers
Download pretrained ASVspoof model of SE-ResNet34
I want to use SE-ResNet34 or SE-ResNet18, SE-ResNet50,... pretrained model's file to serve to generate adversarial samples with PGD attack method. However, I can not find any public pretrained file.
Please help me and give me a link to download…
Mai Bùi
- 21
- 1
- 3
1
vote
0 answers
Generate adversarial patch on GPU
When trying to generate adversarial patch on a GPU server I get this error:
RuntimeError: Expected all tensors to be on the same device, but found
at least two devices, cpu and cuda:0! (when checking argument for
argument tensors in method…
William
- 11
- 2
1
vote
0 answers
TypeError: transpose() received an invalid combination of arguments - got (tuple), but expected one of: (int dim0, int dim1) or (name dim0, name dim1)
I am trying to generate adversarial images using the FastGradientMethod attack in the ART library on a YOLOv5 object detection model. However, I am running into an error when attempting to generate the adversarial image using the fgm.generate()…
Mark Truss
- 11
- 1
1
vote
1 answer
why does pytorch's utils.save_image() change the color of my image
I am saving two images with pytorch's utils.save_image() function here. one is the real image and the other, the perturbed image (just a patch). However, the latter image lost its real appearance when saved with save_image().
# save source…
Craving_gold
- 189
- 1
- 14
1
vote
1 answer
FGSM attack in Foolbox
I am using Foolbox 3.3.1 to perform some adversarial attacks on resnet50 network. The code is as follows:
import torch
from torchvision import models
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model =…
lmsasu
- 7,459
- 18
- 79
- 113
0
votes
0 answers
On-the-fly data augmentation
I would like to train a dataset for images on the fly adversarial. That means I would have to apply data augmentation on the fly and additionally I could integrate different attacks and sampling methods in the training process.
Unfortunately I don't…
Sara
- 1
0
votes
0 answers
The model loss.grad is not None,but the parameters.grad in model is None when training
I'm traing a adversarial net against DQN, the loss uses the Q value of original obs and attacked obs , when training the net, I find the gard_value of parameter is None.How can I resolve it?
for epoch in range(n_epoch):
for…
li xj
- 1
0
votes
0 answers
Negative dimensions error when predicting svm
I have trained an SVM using the NSL-KDD dataset but when I try to make a prediction on it, it comes up with the error "negative dimensions are not allowed".
Below is my code:
`#Indexing of features and labels
features = dataset[:,:-2]
labels =…
james6601
- 1
- 3
0
votes
0 answers
DeepFool attack on MLP provide bad results
I'm trying to attack an MLP with DeepFool, but plotting results I have some strange behavior.
First of all the MLP structure is as follows:
Dense(16, activation='relu', input_shape=(512,)) Dense(16, activation='relu') Dense(2,…
0
votes
0 answers
problems using the ART library with a from scratch trained model
I am a student of artificial intelligence.
I am trying to use the ART library to perform attacks on a binary classifier doing speaker recognition. However, due to some technical problems, I had to resort to training from scratch, using a Keras…
carpeug
- 21
- 3
0
votes
0 answers
AttributeError: 'int' object has no attribute 'dtype' in tensor flow for IBM adversarial robustness library
I want to build a simple adversarial attack using IBM adversarial Library. my code worked before property but after update I am receiving the following error:
AttributeError: 'int' object has no attribute 'dtype'
I have used the example that is…
MSN
- 173
- 4
- 12
0
votes
0 answers
Generating noise to create adversarial images
I have to add noise to MNIST images under the following constraints.
delta is the noise that I am allowed to add, epsilon and alpha can be a free choice.
Can someone tell me how can this be done using numpy?
My attempt:
def…
leech
- 21
- 4
0
votes
1 answer
Why is tape.gradient returning None?
I am trying to create an adversarial example using this function:
def create_adversarial_pattern(input_image, input_label):
input_image = tf.cast(input_image, tf.float32)
with tf.GradientTape() as tape:
tape.watch(input_image)
…
Miles DeBoer
- 7
- 2
0
votes
1 answer
ModuleNotFoundError: No module named 'cleverhans.future'
I'm trying to install cleverhans package on Google Colab and currently getting an error for ModuleNotFoundError: No module named 'cleverhans.future'
Code Attempted:
!pip install -qq -e…
maximus
- 335
- 2
- 16
0
votes
1 answer
How to access the origin label of an image through python script
I am trying to practice adversarial attacks in orange data mining, but am having some trouble loading the image in a python script widget. I have it set up where the import images widget is connected to the python script. 'image' has two labels that…
Miles DeBoer
- 7
- 2