Training issues

[A1ita99]

when I run the following code:

python main.py --mode train --dataset CelebA --celeba_image_dir data/celeba/images --attr_path data/celeba/list_attr_celeba.txt --image_size 128 --c_dim 5 --sample_dir results/samples --log_dir results/logs --model_save_dir results/models --result_dir results/results --selected_attrs Sideburns Bangs Blond_Hair Brown_Hair Male Eyeglasses --lambda_noise 800 --attention True

the generated images in the 'samples' directory appear distorted. How can I fix this?

[bajiua]

当我运行以下代码时：

python main.py --mode train --dataset CelebA --celeba_image_dir data/celeba/images --attr_path data/celeba/list_attr_celeba.txt --image_size 128 --c_dim 5 --sample_dir results/samples --log_dir results/logs --model_save_dir results/models --result_dir results/results --selected_attrs 鬓角 刘海 金发 棕发 男性 眼镜 --lambda_noise 800 --attention True

“samples”目录中生成的图像出现扭曲。我该如何修复？

I met the same problem. Have you solved this problem now?