RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 2 but got size 1 for tensor number 1 in the list.

[gebaltso]

Can anyone help with this error? I used the default setting, didn't change anything.

[Sooplex]

The issue might be as mentioned here: #182 (comment)
You could add the following line in the load_photomaker_adapter() in photomaker/pipeline.py:
self.num_tokens =2 if pm_version =='v2'else 1 ## this seems ignore the compatibility with v1

[gebaltso]

The issue might be as mentioned here: #182 (comment) You could add the following line in the load_photomaker_adapter() in photomaker/pipeline.py: self.num_tokens =2 if pm_version =='v2'else 1 ## this seems ignore the compatibility with v1

Thanks but unfortunately I get the same error.

[Hzzone]

The issue might be as mentioned here: #182 (comment) You could add the following line in the load_photomaker_adapter() in photomaker/pipeline.py: self.num_tokens =2 if pm_version =='v2'else 1 ## this seems ignore the compatibility with v1

Thanks but unfortunately I get the same error.

You have to set the args to use v1:

pipe.load_photomaker_adapter(
    os.path.dirname(photomaker_path),
    subfolder="",
    weight_name=os.path.basename(photomaker_path),
    trigger_word="img",  # define the trigger word
    pm_version='v1',
)    

[gebaltso]

The issue might be as mentioned here: #182 (comment) You could add the following line in the load_photomaker_adapter() in photomaker/pipeline.py: self.num_tokens =2 if pm_version =='v2'else 1 ## this seems ignore the compatibility with v1

Thanks but unfortunately I get the same error.

You have to set the args to use v1:

pipe.load_photomaker_adapter(
    os.path.dirname(photomaker_path),
    subfolder="",
    weight_name=os.path.basename(photomaker_path),
    trigger_word="img",  # define the trigger word
    pm_version='v1',
)    

Thanks but keep getting the same error. I changed the args as you suggested.