Open ArthurConan opened 6 years ago
I am aware of this issue, and will push a fix soon. This is caused by certain tensorflow versions. Thanks for reporting!
@ArthurConan Can you try the latest commit and see if it solves this issue? Thanks!
Hi Huan! Do you have the generated attack images of cifar-10?
Also is it necessary to run the train_model.py before running the test_attack.py? Since I am running out of memory. Thanks a lot!
Yes it is necessary to run train_model.py to train the models before you start the attack. I don't have any images at this time, except for those ones presented in our paper.
Thanks Huan! Do you know is there any online resource that has attack data that is available for downloading?
I tried to execute attack on ImageNet with first example command: python3 test_all.py --untargeted -a black -d imagenet -n 150 --solver adam -b 1 -c 10.0 --use_resize --reset_adam -m 1500 -p 10 -s "imagenet_untargeted"
Can anyone help me with error. Don't know what to do. Thanks in advance.
Traceback (most recent call last): File "test_all.py", line 262, in
main(args)
File "test_all.py", line 102, in main
data, model = ImageNet(), InceptionModel(sess, use_log)
File "/root/Documents/python/ZOO-Attack-master/setup_inception.py", line 223, in init
create_graph()
File "/root/Documents/python/ZOO-Attack-master/setup_inception.py", line 123, in create_graph
FLAGS.model_dir, 'frozen_inception_v3.pb'), 'rb') as f:
File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/platform/flags.py", line 84, in getattr
wrapped(_sys.argv)
File "/usr/local/lib/python3.6/dist-packages/absl/flags/_flagvalues.py", line 630, in call
name, value, suggestions=suggestions)
absl.flags._exceptions.UnrecognizedFlagError: Unknown command line flag 'untargeted'