e4t-diffusion
e4t-diffusion copied to clipboard
{placeholder_token} vs *s in inference
In the fine tune code, there is an assert hard code the special token as {placeholder_tokne}.
"{placeholder_token}" in args.prompt_template
), "You must specify the location of placeholder token by '{placeholder_token}'"
what is the logic we use {placeholder_token} in fine-tuning, but *s in pre-train and inference
{placeholder_token} will be replace with *s when domain-duning process