Skip to content

Mrgrylls11/StarGAN_Watermarking

Repository files navigation

To Train the model the attributes are: python main.py --mode train --dataset CelebA
--image_size 128 --batch_size 1
--num_iters 10000 --model_save_step 2000 --sample_step 500
--log_step 10 --num_workers 2
--apply_watermark True
--watermark_path watermark/watermarks/logo.png
--watermark_size_percent 10
--watermark_position bottom-right
--watermark_output_dir stargan_results/watermarked_results
--save_noisy_intermediate True
--watermark_intermediate_dir stargan_results/intermediate_noisy # Add other parameters like --selected_attrs if needed

About

StarGAN with watermarking.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published