UGATIT
UGATIT copied to clipboard
Official Tensorflow implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation (ICLR 2020)
I want to use Tensorflow 2x because Google colab does not allow Tensorflow1x. Editing the code to run it with TF2 is really difficult for a beginner like me. Hope...
Hi @taki0112 When reading the checkpoint file **_selfie2anime(100 epoch);_** I get tensorflow.python.framework.errors_impl.OutOfRangeError. The tensorflow version I am using is 1.14.0. Can you please guide me on the issue. Below are...
I failed to use the gpu device with 8G memory to run the full UGATIT code. How can I modify the code to use two gpu devices?
Great work! But I have a question about "5.3.2 AdaLIN analysis" in your paper. AdaIN is a model that need a style feature and a content feature, and how to...
The models for 100 epoches are not very effective.If it's convenient for you, can you upload checkpoints after more epoches? Thanks. Best wishes.
只有一个train和test有图片,其他地方没有看到图片呀?
I get using the standard `unzip` program this error message: ``` $ unzip ../50_epoch_selfie2anime_checkpoint.zip Archive: ../50_epoch_selfie2anime_checkpoint.zip warning [../50_epoch_selfie2anime_checkpoint.zip]: 4294967296 extra bytes at beginning or within zipfile (attempting to process anyway)...
So I wanted to see what my daughter looked like in anime but it honestly looked ugly. here is a photo of her:  Anime: 