search

facebookresearch / pytorch_GAN_zoo

A mix of GAN implementations including progressive growing
BSD 3-Clause "New" or "Revised" License
1.62k stars 271 forks source link

How to split generated image? #94

Closed ivder closed 4 years ago

ivder commented 4 years ago

After evaluating the model using python eval.py visualization -n $modelName -m $modelType it will generate a big image consists of nImages = (256 // 2**(max(scale - 2, 3))) * 8 , for example 8x32 = 256 images.

How to split / save the small images separately instead of the combined big image? 1 jpg for 1 generated image

Molugan commented 4 years ago

Hi,

To do that you can add the option --save_dataset $PATH_OUTPUT to your script: it will export 1 image per jpg. The default number of images to export is 1000 you might want to change that.

ivder commented 4 years ago

@Molugan thx for the confirmation. The --save_dataset flag usefulness is a little bit unclear in the Readme file. Would be lot better if you add a sample image of the command's result. Thank you by the way