that Total testing data:250, correct classified data:0
Opened this issue · 0 comments
Hi, I really enjoy your program and I tried to execute zoo attack on ImageNet with other images ( I use the same imagenet data set in ZOO attack Github) , but after converting images from png to numpy file and calculate the 1hot label for each image, I don't able to run zoo attack.
I see that Total testing data:250, correct classified data:0
main(args)
File "main.py", line 177, in main
all_orig_img, all_target_labels, all_orig_labels, all_orig_img_id =
generate_attack_data_set(data, args["num_img"], args["img_offset"],
model, attack_type=args["attack_type"], shift_index=shift_index)
File "main.py", line 106, in generate_attack_data_set
orig_img.append(data.test_data[img_offset+sample_index])
IndexError: index 0 is out of bounds for axis 0 with size 0
Could you plaease help me to solve this issue?
Thanks in advance