tkwoo / anogan-keras Goto Github PK

또 질문 드리네요..

anomaly 이미지를 생성했고 이에 대한 score 를 봤는데요..
anomaly 이미지에서 어떤 부분이 anomaly score 에 높게 반영 되는지 알고 싶습니다.

파란색 부분?? 빨간색 부분?? 아님 다른 곳??

감사합니다..

Hi, @tkwoo

In the case of Anomaly Gan, I knew that it trained with only normal datasets.

At train AnoGAN with MNIST data, Does the train set and test set all contain only normal data?

Or are both normal + abnormal divided into folders?
I wonder.

form @bemoregt.

안녕하세요..

올려주신 소스코드는 잘 동작하는 것을 봤습니다.

몇가지 질문 드리고 싶습니다.

1. 이미지 크기 변경 방법..
2. training 을 label 1만 하셨는데... 2, 3 을 모두 training 하고 나서,
   training 하지 않은 이미지를 test 하였을 때는 anomaly score 가 어떻게 나올까요??

감사합니다.

tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'batch_normalization_1/keras_learning_phase' with dtype bool

Why are we using 'mse' as the loss function for both generator and discrimator? Do we not use 'binary_crossentropy' in case of the optimizers?

Also another doubt was to know the reason behind the usage of Conv2dTranspose layers instead of Upsampling layers?

├── main.py
├── anogan.py
├── weights
├── discriminator.h5
└── generator.h5
└── result
└── save the generated images when training
However, there is no folder of "weights"

Hi,
Thans for your work. I have a question about after train progress. When we send abnoamal image to the model, it can only give a abnomal score, but how can we make sure model classify this image as nomal or abnomal? And I noticed in paper, there are some evaluation index for model ,such as precision,recall,AUC and so on ,how to realize this fuction?
Thanks again for your kind reply.

OSError: Unable to create file (unable to open file: name = 'weights/generator.h5', errno = 2, error message = 'No such file or directory', flags = 13, o_flags = 302)

hello，Is the envrionment of your computer gpu? And i want to know is the code suited the cpu.Thanks for your reply