Open Son-GH opened 1 year ago
Hi, I have a question about the results of the experiment.
I got the best FID 22.4 instead of the reported score 11.92 in paper.
I use the following command. "python train.py --gpus=8 --data=/FFHQ256_2k.zip --cfg=paper256 --outdir=InsG_2k --mirror=1"
And I generated the FFHQ-2k data according to the official styleGAN-ada GitHub. (https://github.com/NVlabs/stylegan2-ada-pytorch)
What should I do for scores on paper?
Hi, I have a question about the results of the experiment.
I got the best FID 22.4 instead of the reported score 11.92 in paper.
I use the following command. "python train.py --gpus=8 --data=/FFHQ256_2k.zip --cfg=paper256 --outdir=InsG_2k --mirror=1"
And I generated the FFHQ-2k data according to the official styleGAN-ada GitHub. (https://github.com/NVlabs/stylegan2-ada-pytorch)
What should I do for scores on paper?