Description: When using wildcards in prompts, each token can only utilize one wildcard. If the token has multiple wildcards (e.g. __haircolor__ __hairstyle__), the following error occurs: WARNING: MainProcess waitress-3 onnx_web.diffusers.utils: unknown wildcard: __haircolor__ __hairstyle__. The image will still continue to generate as normal, there will just be nothing where the wildcards should be. I am pretty sure this is a direct result of the changes made to support dynamic wildcards; however, this occurs with all wildcards, including those from .txt files.
Example:
The image on the right was generated with two wildcards in one token: woman with __billions_of_all_in_one\person\regular\haircolor-unconv__ __billions_of_all_in_one\person\regular\haircuts-female__
The image on the left split the token up into two tokens with one wildcard each: woman with __billions_of_all_in_one\person\regular\haircolor-unconv__ hair, __billions_of_all_in_one\person\regular\haircuts-female__
Description: When using wildcards in prompts, each token can only utilize one wildcard. If the token has multiple wildcards (e.g.
__haircolor__ __hairstyle__
), the following error occurs:WARNING: MainProcess waitress-3 onnx_web.diffusers.utils: unknown wildcard: __haircolor__ __hairstyle__
. The image will still continue to generate as normal, there will just be nothing where the wildcards should be. I am pretty sure this is a direct result of the changes made to support dynamic wildcards; however, this occurs with all wildcards, including those from .txt files.Example:
woman with __billions_of_all_in_one\person\regular\haircolor-unconv__ __billions_of_all_in_one\person\regular\haircuts-female__
woman with __billions_of_all_in_one\person\regular\haircolor-unconv__ hair, __billions_of_all_in_one\person\regular\haircuts-female__