xubo
xubo
> I ran the codes on my computer for python3 and tensorflow-gpu, but I found a trouble that: > stddev=np.sqrt(1.0 / self.latent_dim)), > TypeError: unsupported operand type(s) for /: 'float'...
> Hi,Thanks for your sharing.I run test.py .I encountered a problem and solved it. > > **Error**:AttributeError: 'Blade_runner' object has no attribute 'perturb_mD_img' > line 357 > > ``` >...
> Hi,Thanks for your sharing.I run test.py .I encountered a problem and solved it. > > **Error**:AttributeError: 'Blade_runner' object has no attribute 'perturb_mD_img' > line 357 > > ``` >...
> same problem here. > I've tried to modify the threshold values, increase the batch size, adjusted the learning rate, momentum, increase the epoch, etc.. but to no avail. The...
> I have try to run train.py of the project, and run test.py of the project using the latest patch that saved by train.py (without using the untargetd_DPATCH which download...
> I found the same problem. My solution is to send the copy of the input (input.clone()) when calling the run function. 也许只是眼睛看起来不太明显，但是利用函数求原始图片以及对抗样本的distortion，还是能够看出区别，很明显针对两幅原图比较，结果为0，原始图片与对抗图片肯定不能为0。
> > I found the same problem. My solution is to send the copy of the input (input.clone()) when calling the run function. > > 也许只是眼睛看起来不太明显，但是利用函数求原始图片以及对抗样本的distortion，还是能够看出区别，很明显针对两幅原图比较，结果为0，原始图片与对抗图片肯定不能为0。 > > 其实很明显，你只要在optimizer里不停打印input_adv和input_var就能看出每次更新，input_var也会被更新。我还没细究原因，目前看，可能是直接用input_var的话， pytorch内部机制导致一些关于grad的地方没有真正被关掉。我尝试在使用input_var的地方先clone一份出来，然后发现，input就不变了。...
> > I found the same problem. My solution is to send the copy of the input (input.clone()) when calling the run function. > > 也许只是眼睛看起来不太明显，但是利用函数求原始图片以及对抗样本的distortion，还是能够看出区别，很明显针对两幅原图比较，结果为0，原始图片与对抗图片肯定不能为0。 > > 其实很明显，你只要在optimizer里不停打印input_adv和input_var就能看出每次更新，input_var也会被更新。我还没细究原因，目前看，可能是直接用input_var的话， pytorch内部机制导致一些关于grad的地方没有真正被关掉。我尝试在使用input_var的地方先clone一份出来，然后发现，input就不变了。...