Public/automatic
1
0
Fork 0
mirror of https://github.com/vladmandic/automatic synced 2026-09-17 08:19:11 +02:00
Code Issues Packages Projects Releases Wiki Activity

Ruff fix and refiner->is_refiner.

Browse Source
This commit is contained in:
Hameer Abbasi
2023-08-06 07:37:45 +02:00
parent e65e959eef
commit bb66b5ce0e
3 changed files with 12 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/prompt_parser_diffusers.py
+5 -5
View File
@@ -26,7 +26,7 @@ def compel_encode_prompt(pipeline: typing.Any, *args, **kwargs):
raise TypeError(f"Compel encoding not yet supported for {type(pipeline).__name__}.")
return compel_encode_fn(pipeline, *args, **kwargs)
def compel_encode_prompt_sdxl(pipeline: diffusers.StableDiffusionXLPipeline, prompt: str, negative_prompt: str, prompt_2: typing.Optional[str]=None, negative_prompt_2: typing.Optional[str]=None, refiner=False):
def compel_encode_prompt_sdxl(pipeline: diffusers.StableDiffusionXLPipeline, prompt: str, negative_prompt: str, prompt_2: typing.Optional[str]=None, negative_prompt_2: typing.Optional[str]=None, is_refiner: bool = False):
if shared.opts.data['prompt_attention'] != 'Compel parser':
prompt = convert_to_compel(prompt)
negative_prompt = convert_to_compel(negative_prompt)
@@ -46,20 +46,20 @@ def compel_encode_prompt_sdxl(pipeline: diffusers.StableDiffusionXLPipeline, pro
returned_embeddings_type=ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NON_NORMALIZED,
requires_pooled=True,
)
if refiner is False:
if not is_refiner:
positive_te1 = compel_te1(prompt)
positive_te2, pooled = compel_te2(prompt_2)
positive_te2, positive_pooled = compel_te2(prompt_2)
positive = torch.cat((positive_te1, positive_te2), dim=-1)
negative_te1 = compel_te1(negative_prompt)
negative_te2, negative_pooled = compel_te2(negative_prompt_2)
negative = torch.cat((negative_te1, negative_te2), dim=-1)
else:
positive, pooled = compel_te2(prompt)
positive, positive_pooled = compel_te2(prompt)
negative, negative_pooled = compel_te2(negative_prompt)
shared.log.debug(compel_te1.parse_prompt_string(prompt))
[prompt_embed, negative_embed] = compel_te2.pad_conditioning_tensors_to_same_length([positive, negative])
return prompt_embed, pooled, negative_embed, negative_pooled
return prompt_embed, positive_pooled, negative_embed, negative_pooled
COMPEL_ENCODE_FN_DICT = {diffusers.StableDiffusionXLPipeline: compel_encode_prompt_sdxl}