custom photo: NameError: name 'inverse_image' is not defined

[0width]

Traceback (most recent call last): File "C:\Users\xios.conda\envs\draggan\lib\site-packages\gradio\routes.py", line 406, in run_predict event_data=event_data, File "C:\Users\xios.conda\envs\draggan\lib\site-packages\gradio\blocks.py", line 1300, in process_api fn_index, inputs, iterator, request, event_id, event_data File "C:\Users\xios.conda\envs\draggan\lib\site-packages\gradio\blocks.py", line 1023, in call_function fn, *processed_input, limiter=self.limiter File "C:\Users\xios.conda\envs\draggan\lib\site-packages\anyio\to_thread.py", line 34, in run_sync func, *args, cancellable=cancellable, limiter=limiter File "C:\Users\xios.conda\envs\draggan\lib\site-packages\anyio_backends_asyncio.py", line 877, in run_sync_in_worker_thread return await future File "C:\Users\xios.conda\envs\draggan\lib\site-packages\anyio_backends_asyncio.py", line 807, in run result = context.run(func, *args) File "C:\Users\xios.conda\envs\draggan\lib\site-packages\draggan\web.py", line 183, in on_image_change result = inverse_image( NameError: name 'inverse_image' is not defined

[alienhd]

Same for me.

File "PathtoDragGAN\draggan\web.py", line 183, in on_image_change
    result = inverse_image(
NameError: name 'inverse_image' is not defined

[Zeqiang-Lai]

1.1.5 has a major update which makes old inverse_image incompatible. I am still working on the new one. If you want to try custom image now. Downgrade to 1.1.0

pip install draggan==1.1.0

[tk-da]

Doesn't work with 1.1.0 either: line 183, in on_image_change result = inverse_image( NameError: name 'inverse_image' is not defined

[tk-da]

Same with docker version, maybe this is missing: from .stylegan2.inversion import inverse_image ?

[Zeqiang-Lai]

Same with docker version, maybe this is missing: from .stylegan2.inversion import inverse_image ?

Are you sure you are running via the following commands ?

pip install draggan==1.1.0
python -m draggan.web

[tk-da]

Hm didn't restart my terminal, but now I get AttributeError: module 'torchvision.transforms' has no attribute 'Resize' when trying to load my image.

[Zeqiang-Lai]

See https://github.com/Zeqiang-Lai/DragGAN/issues/39#issuecomment-1594861501

[tk-da]

Thanks!

[OnlySoaring]

Same with docker version, maybe this is missing: from .stylegan2.inversion import inverse_image ?

Are you sure you are running via the following commands ?

pip install draggan==1.1.0
python -m draggan.web

大佬，我draggan退回到1.1.0也还是报这个错唉，如果导入from draggan.deprecated.stylegan2.inversion import inverse_image的话，他会报 File "F:\DragGAN-main\gradio_app.py", line 184, in on_image_change model.g_ema, AttributeError: 'dict' object has no attribute 'g_ema'