StyleSwin
StyleSwin copied to clipboard
Query: How to save the generated samples
Thank you authors for sharing the interesting work you have done in deep generative modeling. I have a rather small doubt, this is regarding the command to generate samples.
python -m torch.distributed.launch --nproc_per_node=1 train_styleswin.py --sample_path /path_to_save_generated_samples --size 256 --G_channel_multiplier 2 --ckpt /path/to/checkpoint --eval --val_num_batches 12500 --val_batch_size 4 --eval_gt_path /path_to_real_images_50k
Here, I wanted to know what should bepath_to_save_generated_samples
?
I am getting the following error message:
Traceback (most recent call last):
File "train_styleswin.py", line 382, in <module>
os.mkdir(args.sample_path)
FileNotFoundError: [Errno 2] No such file or directory: '/content/drive/MyDrive/Repositories/StyleSwin/StyleSwin_generated_samples/samples'
I made a directory in the cloned repository called StyleSwin_generated_samples
for saving the samples.
I changed the /content/drive/MyDrive/Repositories/StyleSwin/StyleSwin_generated_samples/samples
to /content/drive/MyDrive/Repositories/StyleSwin/imgs/
and the command is working.
Additionally, when I try the path /content/drive/MyDrive/Repositories/StyleSwin/StyleSwin_generated_samples/samples/
(with /
at the end of the path) it is still not working
Hi, please make sure your sample_path exist without /samples
added. Thanks.
Thank you for the response.
I wanted to know how to modify /path_to_real_images_50k
. What is the function of this argument? This is the final argument in the commands suggested here.
This augment is used to calculate fid score between the generated image samples and real image samples, which is used in evaluation. If you want to calculate fid score, please put the real images in /path_to_real_images_50k, otherwise, you should comment the fid score calculation in evaluation. Thanks.
Thank you for the response.
I wanted to know how to modify
/path_to_real_images_50k
. What is the function of this argument? This is the final argument in the commands suggested here.
Hi,I encountered the same error as you, have you solved it?