Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Structure of a new training data set #117

Open
DanMenCas opened this issue Apr 6, 2024 · 2 comments
Open

Structure of a new training data set #117

DanMenCas opened this issue Apr 6, 2024 · 2 comments

Comments

@DanMenCas
Copy link

Hi,

How i can create a new data set for use StyleGAN, what should be the structure of this data set?, where i can put the test/validation set and how i should label the images.

Thanks!

@jasuriy
Copy link

jasuriy commented May 13, 2024

@DanMenCas hi
did you resolve your issue or found answer to your question ?

@a-ceron
Copy link

a-ceron commented Oct 10, 2024

Hi, here are some commands that show how you can create a compressed file to use as a dataset:

python dataset_tool.py create_from_images ~/datasets/custom ~/custom-images
python dataset_tool.py display ~/datasets/custom

Then, to train, just use:


python train.py --outdir=~/training-runs --gpus=1 --data=~/datasets/custom --dry-run
python train.py --outdir=~/training-runs --gpus=1 --data=~/datasets/custom

This worked fine for me. I had to modify a few lines, but nothing too complicated.

Also, why do you need test/validation? If you want to use IS or FID, there are some files that show how to pass those datasets, but for training, the previous commands should work.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants