-
UGATIT 预训练模型和数据集百度云下载,为方便谷歌硬盘无法使用的人,不懂的可加Q群857449786 注明UGATIT 共同研究 ,谢谢!
官方已经开放的数据集和模型在google硬盘,比较大,可能无法下载,我分卷压缩了一下,分享到百度云,方便大家下载。
Pretrained model: selfie2anime checkpoint (100 epoch)
Dataset:sel…
-
It is unstable using your newest release code in 64*64 resolution,why?
-
Hi,
Thanks for the great work!
I am going to train according to the guideline you mentioned.
But having couple of issues as below while trying:
1. When run train.py with the full command using…
-
Hi, I enjoyed the paper and was able to run the training example.
I set up the CelebA dataset with 1200 identities provided in legit_indices.npy. However, I am not sure about the hyper-parameter set…
-
Thanks for the nice work!
And we are trying to train the PF-AFN in VITON dataset recently but got the below error in train_PBAFN_stage1.
"RuntimeError: Given input size: (512x2x1). Calculated output…
-
Traceback (most recent call last):
File "train_PBAFN_stage1.py", line 30, in
Traceback (most recent call last):
File "train_PBAFN_stage1.py", line 30, in
torch.cuda.set_device(opt.local_rank)
Fil…
-
https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/c028d9f64c64ba50b64348129a6ed2eefdfaffee/models/networks.py#L271-L306
Maybe in pix2pix_model and cycle_gan_model ? Otherwise setting `ga…
-
hello! Can you tell me how to train this model? I meet a issues
train.py [-h] [--name NAME] [--model MODEL] [--mask_type MASK_TYPE]
[--checkpoints_dir CHECKPOINTS_DIR] [--which…
-
@tcwang0509 @junyanz @mingyuliutw @elfprince13 @javl @ufoym @borisfom @TheAussieStew
Hello everyone. I am unable to completely understand and interpret the tensorboard loss curves produced by this m…
-
## 一言でいうと
GANにおいて、Discriminatorは真偽を判定するためその出力は確率値(sigmoid)になっている。ただ、これだと勾配消失が起きやすい。そこで二乗誤差を使用する手法を提案。具体的には、ネットワークの出力値から定数を引く形で誤差を定義する。この最小化が、ピアソンのカイ二乗分布の分散の最小化と等価であることも証明。
### 論文リンク
https://a…