Public/automatic
1
0
Fork 0
mirror of https://github.com/vladmandic/automatic synced 2026-09-19 09:14:35 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix compel to full and add batch sizes

Browse Source
This commit is contained in:
Vladimir Mandic
2023-08-09 10:28:44 +00:00
parent e2b0d981ac
commit 16725ab38f
2 changed files with 58 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/processing_diffusers.py
+16 -10
View File
@@ -53,7 +53,7 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
return imgs
def set_pipeline_args(model, prompt: str, negative_prompt: str, prompt_2: typing.Optional[str] =None, negative_prompt_2: typing.Optional[str] = None, is_refiner: bool = False, **kwargs):
def set_pipeline_args(model, prompts: list, negative_prompts: list, prompts_2: typing.Optional[list]=None, negative_prompts_2: typing.Optional[list]=None, is_refiner: bool=False, **kwargs):
args = {}
pipeline = model
signature = inspect.signature(type(pipeline).__call__)
@@ -65,7 +65,13 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
negative_embed = None
negative_pooled = None
if shared.opts.data['prompt_attention'] in {'Compel parser', 'Full parser'}:
prompt_embed, pooled, negative_embed, negative_pooled = prompt_parser_diffusers.compel_encode_prompt(model, prompt, negative_prompt, prompt_2, negative_prompt_2, is_refiner, kwargs.pop("clip_skip", None))
prompt_embed, pooled, negative_embed, negative_pooled = prompt_parser_diffusers.compel_encode_prompts(model,
prompts,
negative_prompts,
prompts_2,
negative_prompts_2,
is_refiner,
kwargs.pop("clip_skip", None))
if 'prompt' in possible:
if hasattr(model, 'text_encoder') and 'prompt_embeds' in possible and prompt_embed is not None:
args['prompt_embeds'] = prompt_embed
@@ -73,7 +79,7 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
args['pooled_prompt_embeds'] = pooled
args['prompt_2'] = None #Cannot pass prompts when passing embeds
else:
args['prompt'] = prompt
args['prompt'] = prompts
if 'negative_prompt' in possible:
if hasattr(model, 'text_encoder') and 'negative_prompt_embeds' in possible and negative_embed is not None:
args['negative_prompt_embeds'] = negative_embed
@@ -81,7 +87,7 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
args['negative_pooled_prompt_embeds'] = negative_pooled
args['negative_prompt_2'] = None
else:
args['negative_prompt'] = negative_prompt
args['negative_prompt'] = negative_prompts
if 'num_inference_steps' in possible:
args['num_inference_steps'] = p.steps
if 'guidance_scale' in possible:
@@ -157,10 +163,10 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
refiner_enabled = shared.sd_refiner is not None and p.enable_hr
pipe_args = set_pipeline_args(
model=shared.sd_model,
prompt=prompts,
negative_prompt=negative_prompts,
prompt_2=[p.refiner_prompt] if len(p.refiner_prompt) > 0 else prompts,
negative_prompt_2=[p.refiner_negative] if len(p.refiner_negative) > 0 else negative_prompts,
prompts=prompts,
negative_prompts=negative_prompts,
prompts_2=[p.refiner_prompt] if len(p.refiner_prompt) > 0 else prompts,
negative_prompts_2=[p.refiner_negative] if len(p.refiner_negative) > 0 else negative_prompts,
eta=shared.opts.eta_ddim,
guidance_rescale=p.diffusers_guidance_rescale,
denoising_start=0 if refiner_enabled and p.refiner_start > 0 and p.refiner_start < 1 else None,
@@ -211,8 +217,8 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
for i in range(len(output.images)):
pipe_args = set_pipeline_args(
model=shared.sd_refiner,
prompt=[p.refiner_prompt] if len(p.refiner_prompt) > 0 else prompts[i],
negative_prompt=[p.refiner_negative] if len(p.refiner_negative) > 0 else negative_prompts[i],
prompts=[p.refiner_prompt] if len(p.refiner_prompt) > 0 else prompts[i],
negative_prompts=[p.refiner_negative] if len(p.refiner_negative) > 0 else negative_prompts[i],
num_inference_steps=p.hr_second_pass_steps,
eta=shared.opts.eta_ddim,
strength=p.denoising_strength,
modules/prompt_parser_diffusers.py
+42 -28
View File
@@ -1,3 +1,4 @@
import os
import typing
import torch
import diffusers
@@ -5,13 +6,14 @@ from compel import Compel, ReturnedEmbeddingsType
import modules.shared as shared
import modules.prompt_parser as prompt_parser
debug_output = os.environ.get('SD_PROMPT_DEBUG', None)
debug = shared.log.info if debug_output is not None else lambda *args, **kwargs: None
def convert_to_compel(prompt: str):
if prompt is None:
return None
all_schedules = prompt_parser.get_learned_conditioning_prompt_schedules(
prompt, 100
)[0]
all_schedules = prompt_parser.get_learned_conditioning_prompt_schedules([prompt], 100)[0]
output_list = prompt_parser.parse_prompt_attention(all_schedules[0][1])
converted_prompt = []
for subprompt, weight in output_list:
@@ -31,8 +33,35 @@ CLIP_SKIP_MAPPING = {
}
def compel_encode_prompts(
pipeline: diffusers.StableDiffusionXLPipeline | diffusers.StableDiffusionPipeline,
prompts: list,
negative_prompts: list,
prompts_2: typing.Optional[list] = None,
negative_prompts_2: typing.Optional[list] = None,
is_refiner: bool = None,
clip_skip: typing.Optional[int] = None,
):
prompt_embeds = []
positive_pooleds = []
negative_embeds = []
negative_pooleds = []
for i in range(len(prompts)):
prompt_embed, positive_pooled, negative_embed, negative_pooled = compel_encode_prompt(pipeline, prompts[i], negative_prompts[i], prompts_2[i], negative_prompts_2[i], is_refiner, clip_skip)
prompt_embeds.append(prompt_embed)
positive_pooleds.append(positive_pooled)
negative_embeds.append(negative_embed)
negative_pooleds.append(negative_pooled)
prompt_embeds = torch.cat(prompt_embeds, dim=0)
positive_pooleds = torch.cat(positive_pooleds, dim=0)
negative_embeds = torch.cat(negative_embeds, dim=0)
negative_pooleds = torch.cat(negative_pooleds, dim=0)
return prompt_embeds, positive_pooleds, negative_embeds, negative_pooleds
def compel_encode_prompt(
pipeline: diffusers.StableDiffusionXLPipeline,
pipeline: diffusers.StableDiffusionXLPipeline | diffusers.StableDiffusionPipeline,
prompt: str,
negative_prompt: str,
prompt_2: typing.Optional[str] = None,
@@ -41,24 +70,17 @@ def compel_encode_prompt(
clip_skip: typing.Optional[int] = None,
):
if shared.sd_model_type not in {"sd", "sdxl"}:
shared.log.warning(
f"Compel encoding not yet supported for {type(pipeline).__name__}."
)
shared.log.warning(f"Prompt parser: Compel not supported: {type(pipeline).__name__}")
return (None, None, None, None)
if shared.sd_model_type == "sdxl":
embedding_type = ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NON_NORMALIZED
if clip_skip is not None:
shared.log.debug("CLIP skip ignored as it is unsupported for SDXL")
if clip_skip is not None and clip_skip > 1:
shared.log.warning(f"Prompt parser SDXL unsupported: clip_skip={clip_skip}")
else:
embedding_type = CLIP_SKIP_MAPPING.get(
clip_skip, ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NORMALIZED
)
embedding_type = CLIP_SKIP_MAPPING.get(clip_skip, ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NORMALIZED)
if clip_skip not in CLIP_SKIP_MAPPING:
shared.log.warning(
f"Recieved a CLIP skip of {clip_skip}, but only {set(CLIP_SKIP_MAPPING.keys())} is supported. "
"Falling back to 2."
)
shared.log.warning(f"Prompt parser unsupported: clip_skip={clip_skip} expected={set(CLIP_SKIP_MAPPING.keys())}")
if shared.opts.data["prompt_attention"] != "Compel parser":
prompt = convert_to_compel(prompt)
@@ -84,25 +106,17 @@ def compel_encode_prompt(
positive_te1 = compel_te1(prompt)
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, positive_pooled = compel_te2(prompt)
negative, negative_pooled = compel_te2(negative_prompt)
shared.log.debug(
f"Parsed Compel string: {compel_te1.parse_prompt_string(prompt)}"
)
[
prompt_embed,
negative_embed,
] = compel_te2.pad_conditioning_tensors_to_same_length([positive, negative])
parsed = compel_te1.parse_prompt_string(prompt)
debug(f"Prompt parser Compel: {parsed}")
[prompt_embed, negative_embed] = compel_te2.pad_conditioning_tensors_to_same_length([positive, negative])
return prompt_embed, positive_pooled, negative_embed, negative_pooled
positive, negative = compel_te1(prompt), compel_te1(negative_prompt)
[prompt_embed, negative_embed] = compel_te1.pad_conditioning_tensors_to_same_length(
[positive, negative]
)
[prompt_embed, negative_embed] = compel_te1.pad_conditioning_tensors_to_same_length([positive, negative])
return prompt_embed, None, negative_embed, None