-
I'm running the model and every time it happen the same, when I reach epoch 200 it stops creating checkpoints but nothing is shown in Terminal. I've tried to use ¨--model cycle_gan --epoch_count 200 -…
-
Hello,
I tried to run the gan_mnist.py file both with the most current master version of pytorch (0.2.0+75bb50b) and with an older commit (0.2.0+c62490b).
With both versions the memory used by t…
-
Hi @darkstar112358,
this is a great idea extending the cycle GAN! Is there any chance you'll make it CPU friendly?
When I try to run the PyTorch cycleGAN on the CPU of my 16gb laptop it blows …
ghost updated
7 years ago
-
Hi!
I'm not very clear that why the scheme I can only generate identical images on the horse to zebra translation task. Is it because the same attention network from x to y and from y to x ? Or is …
-
Hi, thanks for your great work. I have trained my datasets with cycle_gan model for 400 epoch. I have generated results with different epochs and qualitatively they have been similar since the 100 ep…
-
Need tips/advice on how best to reduce the unwanted border effects from CycleGAN. See output image examples:
![epoch200_fake_B](https://user-images.githubusercontent.com/26606744/208502979-913eb6a0-0…
-
I have managed to turn dropout on for both pix2pix and cycle gan during training and in inference.
I would now like to explore the impact different values for dropout has on the predictions that a…
-
Hello all,
first of all as big thank you to people like @junyanz for publishing and maintaining this repo.
I am running some experiments with CycleGAN aiming to generate IR/thermal images from R…
-
Hello,
I am trying to run cycle GAN model on grey scale medical image dataset. In the start of the training I get salt and pepper noise which tends to reduce around epoch 40 ,but it is still visibl…
-
## Review
### 一.用[Cycle-GAN](https://arxiv.org/abs/1703.10593)将 SKU的理想图transfer为真实图
#### 符号
`A` : 理想环境数据(摆拍图)
`B` : 真实环境数据(货架图)
`A_r`,`B_r` : real 的数据
`A_f`,`B_f` : fake 的数据
`G` : G(A_r…