search

genforce / idinvert

[ECCV 2020] In-Domain GAN Inversion for Real Image Editing
https://genforce.github.io/idinvert/
MIT License
461 stars 65 forks source link

How to split dataset into a training set, and a test set #19

Closed kkang831 closed 4 years ago

kkang831 commented 4 years ago

Thank you for sharing the code.

To learn the encoder for ffhq dataset, I want to know how to split the dataset, and is there a code for split??

And, also, is there any pre processing for ffhq?

Thank you

zhujiapeng commented 4 years ago

We referred to the dataset_tool.py offered by stylegan. We first get the .png images from .tfrecords file and then save those .png images to the train .tfrecords and the test .tfrecords