search

clovaai / stargan-v2

StarGAN v2 - Official PyTorch Implementation (CVPR 2020)
Other
3.48k stars 656 forks source link

Is there any rule for number of image data, or validation set size when use custom dataset? #42

Open iPsych opened 4 years ago

iPsych commented 4 years ago

I re-runned the training with provided dataset and training code, and guess the previous errors are due to mismatch some 'number' between my custom dataset and AFHQ or CELEBA.

Is there any mandatory fixed number of files in each modalities of val folder, or representative folder? I matched many numbers (number of domains, image size, etc... ) but the numbers of images are quite small (train : 4-600 per domain, val : 100 per each domain), and the images in representative folders are also smaller than examples.

engineer-38 commented 4 years ago

How you arrange the dataset? I'm trying to use a custom dataset, but I do not know how to arrange it inside the "data" file. can you please help me in this.

yunjey commented 4 years ago

@iPsych I think the number of training data should be greater than 10K.

Nick-Sun1018 commented 3 years ago

How you arrange the dataset? I'm trying to use a custom dataset, but I do not know how to arrange it inside the "data" file. can you please help me in this.

Have you solved this issue? I meet the same one.

Zhenggen-deng commented 3 years ago

How you arrange the dataset? I'm trying to use a custom dataset, but I do not know how to arrange it inside the "data" file. can you please help me in this.

What I want to do is the same as you, and I have the same confusion as you. Have you finally solved this problem? Can I add your contact information to communicate with you?