2022-10-16 14:54:33 +00:00
2022-05-25 19:54:48 +02:00
2022-10-07 21:58:18 +00:00
2022-09-30 21:55:50 +00:00
2022-10-12 21:06:10 +00:00
2022-10-12 21:05:33 +00:00
2022-08-06 12:27:24 +00:00
2022-05-25 19:54:48 +02:00
2022-10-12 21:05:49 +00:00
2022-05-25 19:54:48 +02:00
2022-08-06 12:27:24 +00:00
2022-06-16 15:28:52 +00:00
2022-10-02 13:42:04 +00:00
2022-10-16 14:54:33 +00:00
2022-10-16 14:54:33 +00:00
2022-10-16 14:54:33 +00:00
2022-09-25 08:31:13 +00:00
2022-09-30 21:53:17 +00:00
2022-08-06 12:27:24 +00:00
2022-05-25 19:54:48 +02:00
2022-10-16 14:54:33 +00:00
2022-10-16 14:54:33 +00:00
2022-10-16 14:54:33 +00:00
2022-10-16 14:54:33 +00:00
2022-10-16 14:54:33 +00:00
2022-10-16 14:54:33 +00:00
2022-05-25 19:54:48 +02:00
2022-10-12 21:07:00 +00:00
2022-09-30 21:54:29 +00:00
2022-05-25 19:54:48 +02:00
2022-09-25 08:31:13 +00:00
2022-10-07 21:58:48 +00:00
2022-09-25 08:31:13 +00:00
2022-05-25 19:54:48 +02:00
2022-05-25 19:54:48 +02:00
2022-06-16 16:45:21 +00:00
2022-05-25 19:54:48 +02:00
2022-05-25 19:54:48 +02:00
2022-08-06 12:27:24 +00:00
2022-09-30 21:53:17 +00:00
2022-10-07 21:59:07 +00:00
2022-09-30 21:56:38 +00:00
2022-05-25 19:54:48 +02:00
2022-05-25 19:54:48 +02:00
2022-06-17 21:42:45 +02:00
2022-05-25 19:54:48 +02:00
2022-09-25 08:31:13 +00:00
2022-09-22 16:32:37 +00:00
2022-05-25 19:54:48 +02:00
2022-05-25 19:54:48 +02:00
2022-05-25 19:54:48 +02:00
2022-08-06 12:27:24 +00:00
2022-10-07 21:56:23 +00:00
2022-09-30 21:56:08 +00:00
2022-05-25 19:54:48 +02:00
2022-09-25 08:31:13 +00:00
2022-09-27 21:24:41 +00:00
2022-09-13 09:23:01 +00:00
2022-09-16 16:47:07 +00:00
2022-09-02 14:15:54 +00:00
2022-09-02 14:15:54 +00:00
2022-09-25 08:31:13 +00:00
2022-08-06 12:27:24 +00:00

stylegan2-tf-2.x

  • This is tensorflow 2.0 based keras subclassing reimplementation of official StyleGAN2 Repo

Inference from official weights

Official repo result Official weight copied result from this repo
official_result restored_result

Training on 256x256

  • batch size 32
  • around 300k steps
d_loss g_loss
256x256_d_loss 256x256_g_loss
  • result samples
256x256_result_0 256x256_result_1 256x256_result_2

Using custom data

  • Create dataset(tf.data.Dataset) loader which returns
    • shape: [None, 3, res, res] where res is desired image resolution
    • image value range: [-1, +1]

Etc

Using with pycharm

  • If .bashrc file's paths don't work in pycharm environment (e.g. nvcc -h), open pycharm with following
# move to pycharm installed location (location may vary)
moono@moono-ubuntu:~$ cd .local/share/JetBrains/Toolbox/apps/PyCharm-P/ch-0/201.7846.77/bin/

# launch pycharm with .bashrc 
moono@moono-ubuntu:~/.local/share/JetBrains/Toolbox/apps/PyCharm-P/ch-0/201.7846.77/bin$ bash pycharm.sh
S
Description
No description provided
Readme
224 MiB
Languages
Python 97.8%
Cuda 2.1%