taki0112 / cartoongan-tensorflow Goto Github PK

I cannot open the link for the pretrained vgg19 model!

The batch size has been changed to 4. I also changed the iteration. I prepared 4000 images for training. However, converting an image does not look like a cartoon.

Where should I change the other distributed programs?

Hi,

Thanks for this work. However, I do not find the edge_smooth.py and other relevant .py files in the repository.

Thanks,
Akilesh

Hi , thanks for your work, I have trained one model with image size 256*256, I could get good results in samples, but when I test it with other input shape, it get worse , do you know why it happened?

你好作者，那个 vgg19.npy下载不了，怎么解决？
谢谢

Hi, I train the model on P40 GPU about two days, but get a incorrect results.
TrainA is cartoon images, about 5000+.
TrainB is real world video images, about 7000+
My hyperparameter:
batch_size = 1
epoch = 100
init_epoch=10
When finish the train, d_loss: 0.00018069, g_loss: 237.02038574

There hyperparameter are right?
Can you share the train sets info or hyperparameter?

Thank you!

After running the picture, it did not turn into a cartoon picture, but I felt that the resolution was lowered. I don’t know what caused it.

Hi, thank you for your great work,
May I ask you 2 questions about this:

• What is the batch_size during your training?
• How many images in trainA and trainB that you using in your dataset.
  Thank you so much.