_pickle.UnpicklingError: invalid load key, '\xff'.

[nickyvdz-art]

no problems, fixed. - thank you so very much for this, this is truly excellent, what a package!

[nickyvdz-art]

(Wang) C:\Users\Bob\URST\Johnson2016Perceptual>python test.py --content C:\Users\Bob\Desktop\ds.jpg --model C:\Users\Bob\Desktop\dsstyle.jpg --URST
Namespace(URST=True, content='C:\Users\Bob\Desktop\ds.jpg', device='cuda', model='C:\Users\Bob\Desktop\dsstyle.jpg', outf='output', padding=32, patch_size=1000, test_speed=False, thumb_size=1024)
Traceback (most recent call last):
File "test.py", line 76, in
checkpoint = torch.load(args.model)
File "C:\Users\Bob\anaconda3\envs\Wang\lib\site-packages\torch\serialization.py", line 387, in load
return _load(f, map_location, pickle_module, **pickle_load_args)
File "C:\Users\Bob\anaconda3\envs\Wang\lib\site-packages\torch\serialization.py", line 564, in _load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, '\xff'.

this im having trouble working out though?

[Zhe Chen]

(Wang) C:\Users\Bob\URST\Johnson2016Perceptual>python test.py --content C:\Users\Bob\Desktop\ds.jpg --model C:\Users\Bob\Desktop\dsstyle.jpg --URST
Namespace(URST=True, content='C:\Users\Bob\Desktop\ds.jpg', device='cuda', model='C:\Users\Bob\Desktop\dsstyle.jpg', outf='output', padding=32, patch_size=1000, test_speed=False, thumb_size=1024)
Traceback (most recent call last):
File "test.py", line 76, in
checkpoint = torch.load(args.model)
File "C:\Users\Bob\anaconda3\envs\Wang\lib\site-packages\torch\serialization.py", line 387, in load
return _load(f, map_location, pickle_module, **pickle_load_args)
File "C:\Users\Bob\anaconda3\envs\Wang\lib\site-packages\torch\serialization.py", line 564, in _load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, '\xff'.

this im having trouble working out though?

Hi, Johnson2016Perceptual is a style-specific style transfer method, that is, it requires training a model for each style. So this method has no --style option. The --model option requires a .pth model instead of an image.

[nickyvdz-art]

apologies for my misunderstanding, I must have previously used different variation of Johnson2016Perceptual - thanks for getting back to me so soon again, I am really grateful for this, thank you for your work / efforts

…

On Sat, Apr 3, 2021, at 13:41, czczup wrote: > (Wang) C:\Users\Bob\URST\Johnson2016Perceptual>python test.py --content C:\Users\Bob\Desktop\ds.jpg --model C:\Users\Bob\Desktop\dsstyle.jpg --URST > Namespace(URST=True, content='C:\Users\Bob\Desktop\ds.jpg', device='cuda', model='C:\Users\Bob\Desktop\dsstyle.jpg', outf='output', padding=32, patch_size=1000, test_speed=False, thumb_size=1024) > Traceback (most recent call last): > File "test.py", line 76, in > checkpoint = torch.load(args.model) > File "C:\Users\Bob\anaconda3\envs\Wang\lib\site-packages\torch\serialization.py", line 387, in load > return _load(f, map_location, pickle_module, **pickle_load_args) > File "C:\Users\Bob\anaconda3\envs\Wang\lib\site-packages\torch\serialization.py", line 564, in _load > magic_number = pickle_module.load(f, **pickle_load_args) > _pickle.UnpicklingError: invalid load key, '\xff'. > this im having trouble working out though? Hi, Johnson2016Perceptual is a style-specific style transfer method, that is, it requires training a model for each style. So this method has no `--style` option. The `--model` option requires a `.pth` model instead of an image. — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQLUPAHQF67M3HA6YEWWPPTTGZ557ANCNFSM42I6G6PQ>.