fenhua / detdak Goto Github PK

请问一下这个该怎么办呢。改成from mmcv.runner import ...也不太行

File "/faster_rcnn/DeDark/attack_rcnn_mask.py", line 31, in
_, noise = L2_attack(model, img, max_iter=100, epsilon=100, mask=mask) # L2 20->1
File "/faster_rcnn/DeDark/attack_utils/attackloss_rcnn.py", line 102, in L2_attack
idx = GetObject(model, img)
File "/faster_rcnn/DeDark/attack_utils/attackloss_rcnn.py", line 26, in GetObject
if(result[i][j,4] > 0.3):
TypeError: list indices must be integers or slices, not tuple

How can I solve this problem? Looking forward to your reply!

Hi there,

What commands to I have to run to start the attack and do I have to add any parameters to the command for a specific task?
Thanks @FenHua

@FenHua
Hello, I have studied your paper. Is there any code for Heatmap-based Algorithm in this project?

Have you open sourced this code?

Hi, you mentioned in the article that we find that the sensitive locations of the input image
might change over time when attacking algorithms are performed
iteratively, how you proved it？
Thanks!@FenHua

cfgfile = "models/yolov4.cfg"
weightfile = "models/yolov4.weights"

Where did you train your yolo model?

can you share the datasets