How should the filelist look like?

[Cristy94]

First I want to say congratulations for the amazing job you did!

I did not find any example of what the Prepare training images filelist. step consists on. How must the images be listed? What is the format? One file name per line, like this?

pic-1.jpg
pic-2.jpg

[JiahuiYu]

Please refer to #15 .

[Cristy94]

Thank you, I just saw that closed issue, you might want to add a link from the readme to it.

Why are there two lines in the dataset?

  celebahq: [
    'data/celeba_hq/train_shuffled.flist',
    'data/celeba_hq/train_shuffled.flist'
  ]

[JiahuiYu]

Sorry, typo. The second flist is for validation. I've made an update. Thanks for info.

[Cristy94]

So I have to split myself the dataset into test/validation? What does static_view mean?

[JiahuiYu]

static_view is for tensorboard. If static_view is true, same masks (otherwise random masks) will be used for validation images.

[Cristy94]

Cool, thanks a lot. Can I disable static view from the .yml and have random masks instead? I only see a STATIC_VIEW_SIZE: 30 setting.

[JiahuiYu]

You can modify lines.

[lijhong]

Is the training set and validation set the same picture?

[JiahuiYu]

They are not.

[Cristy94]

The validation set is the one used for the network loss output?