thaoshibe / beautygan-pytorch-reimplementation Goto Github PK

Excuse, what is the difference between the setting of args.model == 'cycleGAN' and args.model =='makeupGAN'

FYI.

The link to the dataset can not be used, can you please upload the dataset or a link that can help?

Hi，As you can see, in my test results, it will display non-makeup and makeup images, but other fake_A\fake_B and rec_A\rec_B images are all black. I don't understand why. Can you give me some help

Could you please update URL for Makeup Transfer Dataset? Former URL is not working.

Hi, I used your code to train a new model, however, I don't know how should I use it to generate fake images. Thanks a lot for helping.

histogram_matching has cpu().numpy(), how about training time?

Hello, do you have a plan to release the testing code?

Hi Thaoshibe,
I try to train network by:

python train.py --data_path /path/to/dataset --epochs=5000 --vis_step=100 --task_name=ganmakeup002

I already trained 368 epochs from 5 000 but I am not able to find any output in these subdirectories /vizualization/_ganmakeup002 , /snapshot/_ganmakeup002, /logs/_ganmakeup002. I expected some outputs every 100 epochs (--vis_step=100). Did I miss some parameter?

因为源数据集已经不可用，我在网上找到了数据集但是可能里面的文件和元数据及不太一样，能否提供源数据集的文件夹结构？

When I tried to run this code, it reported such an error：
Traceback (most recent call last):
File "train.py", line 76, in
train_net()
File "train.py", line 59, in train_net
solver.train()
File "/data/home/scv0004/BeautyGAN-pytorch-reimplementation/solver_makeup.py", line 264, in train
for self.i, (img_A, img_B, mask_A, mask_B) in enumerate(tqdm(self.data_loader_train)):
File "/data/home/scv0004/.conda/envs/beautygan/lib/python3.7/site-packages/tqdm/std.py", line 1178, in iter
for obj in iterable:
File "/data/home/scv0004/.conda/envs/beautygan/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 517, in next
data = self._next_data()
File "/data/home/scv0004/.conda/envs/beautygan/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 557, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/data/home/scv0004/.conda/envs/beautygan/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/data/home/scv0004/.conda/envs/beautygan/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/data/home/scv0004/BeautyGAN-pytorch-reimplementation/data_loaders/makeup.py", line 29, in getitem
idxA = random.choice(range(self.noiA))
File "/data/home/scv0004/.conda/envs/beautygan/lib/python3.7/random.py", line 261, in choice
raise IndexError('Cannot choose from an empty sequence') from None
IndexError: Cannot choose from an empty sequence
Can you take a look at it for me when you have time？

Hi, would it be possible to get some sample results from the last/best training epoch or checkpoints? The link to checkpoints that is in the readme does not work :(

Hi,
I have different makeup image database (non-makup + makup images), but there is no segmentation. Could you please advice me how to prepare segmented images? Is there any library?
Thanks