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

Add refiner steps

Browse Source
This commit is contained in:
Disty0
2023-08-27 17:56:55 +03:00
parent 8e6b80104d
commit f2e6e0e88e
5 changed files with 26 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/img2img.py
+2 -1
View File
@@ -77,7 +77,7 @@ def process_batch(p, input_files, input_dir, output_dir, inpaint_mask_dir, args)
shared.log.debug(f'Processed: {len(image_files)} Memory: {memory_stats()} batch')
def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_styles, init_img, sketch, init_img_with_mask, inpaint_color_sketch, inpaint_color_sketch_orig, init_img_inpaint, init_mask_inpaint, steps: int, sampler_index: int, latent_index: int, mask_blur: int, mask_alpha: float, inpainting_fill: int, full_quality: bool, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, diffusers_guidance_rescale: float, refiner_start: float, clip_skip: int, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, selected_scale_tab: int, height: int, width: int, scale_by: float, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_files: list, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, *args): # pylint: disable=unused-argument
def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_styles, init_img, sketch, init_img_with_mask, inpaint_color_sketch, inpaint_color_sketch_orig, init_img_inpaint, init_mask_inpaint, steps: int, sampler_index: int, latent_index: int, mask_blur: int, mask_alpha: float, inpainting_fill: int, full_quality: bool, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, diffusers_guidance_rescale: float, refiner_steps: int, refiner_start: float, clip_skip: int, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, selected_scale_tab: int, height: int, width: int, scale_by: float, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_files: list, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, *args): # pylint: disable=unused-argument
if shared.sd_model is None:
shared.log.warning('Model not loaded')
@@ -171,6 +171,7 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s
denoising_strength=denoising_strength,
image_cfg_scale=image_cfg_scale,
diffusers_guidance_rescale=diffusers_guidance_rescale,
refiner_steps=refiner_steps,
refiner_start=refiner_start,
inpaint_full_res=inpaint_full_res,
inpaint_full_res_padding=inpaint_full_res_padding,
modules/processing.py
+5 -2
View File
@@ -151,6 +151,7 @@ class StableDiffusionProcessing:
self.iteration = 0
self.is_hr_pass = False
self.enable_hr = None
self.refiner_steps = 4
self.refiner_start = 0
self.ops = []
shared.opts.data['clip_skip'] = clip_skip
@@ -851,7 +852,7 @@ def old_hires_fix_first_pass_dimensions(width, height):
class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
sampler = None
def __init__(self, enable_hr: bool = False, denoising_strength: float = 0.75, firstphase_width: int = 0, firstphase_height: int = 0, hr_scale: float = 2.0, hr_upscaler: str = None, hr_second_pass_steps: int = 0, hr_resize_x: int = 0, hr_resize_y: int = 0, refiner_start: float = 0, refiner_prompt: str = '', refiner_negative: str = '', **kwargs):
def __init__(self, enable_hr: bool = False, denoising_strength: float = 0.75, firstphase_width: int = 0, firstphase_height: int = 0, hr_scale: float = 2.0, hr_upscaler: str = None, hr_second_pass_steps: int = 0, hr_resize_x: int = 0, hr_resize_y: int = 0, refiner_steps: int = 4, refiner_start: float = 0, refiner_prompt: str = '', refiner_negative: str = '', **kwargs):
super().__init__(**kwargs)
self.enable_hr = enable_hr
@@ -871,6 +872,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
self.truncate_x = 0
self.truncate_y = 0
self.applied_old_hires_behavior_to = None
self.refiner_steps = refiner_steps
self.refiner_start = refiner_start
self.refiner_prompt = refiner_prompt
self.refiner_negative = refiner_negative
@@ -1018,7 +1020,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing):
class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
sampler = None
def __init__(self, init_images: list = None, resize_mode: int = 0, denoising_strength: float = 0.3, image_cfg_scale: float = None, mask: Any = None, mask_blur: int = 4, inpainting_fill: int = 0, inpaint_full_res: bool = True, inpaint_full_res_padding: int = 0, inpainting_mask_invert: int = 0, initial_noise_multiplier: float = None, refiner_start: float = 0, refiner_prompt: str = '', refiner_negative: str = '', **kwargs):
def __init__(self, init_images: list = None, resize_mode: int = 0, denoising_strength: float = 0.3, image_cfg_scale: float = None, mask: Any = None, mask_blur: int = 4, inpainting_fill: int = 0, inpaint_full_res: bool = True, inpaint_full_res_padding: int = 0, inpainting_mask_invert: int = 0, initial_noise_multiplier: float = None, refiner_steps: int = 4, refiner_start: float = 0, refiner_prompt: str = '', refiner_negative: str = '', **kwargs):
super().__init__(**kwargs)
self.init_images = init_images
self.resize_mode: int = resize_mode
@@ -1037,6 +1039,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
self.mask = None
self.nmask = None
self.image_conditioning = None
self.refiner_steps = refiner_steps
self.refiner_start = refiner_start
self.refiner_prompt = refiner_prompt
self.refiner_negative = refiner_negative
modules/processing_diffusers.py
+8 -3
View File
@@ -261,15 +261,19 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
if shared.opts.diffusers_move_base and not shared.sd_model.has_accelerate:
shared.sd_model.to(devices.device)
use_denoise_start = (is_refiner_enabled and not p.is_hr_pass and p.refiner_start > 0 and p.refiner_start < 1)
base_args = set_pipeline_args(
model=shared.sd_model,
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,
num_inference_steps=p.steps,
num_inference_steps=int(p.steps // (p.refiner_start if use_denoise_start else 1) + (1 if use_denoise_start else 0)),
eta=shared.opts.eta_ddim,
guidance_rescale=p.diffusers_guidance_rescale,
denoising_start=0 if use_denoise_start else None,
denoising_end=p.refiner_start if use_denoise_start else None,
output_type='latent' if hasattr(shared.sd_model, 'vae') else 'np',
is_refiner=False,
clip_skip=p.clip_skip,
@@ -303,7 +307,7 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
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,
num_inference_steps=int((p.hr_second_pass_steps // p.denoising_strength) + 1),
num_inference_steps=int(p.hr_second_pass_steps // p.denoising_strength + 1),
eta=shared.opts.eta_ddim,
guidance_rescale=p.diffusers_guidance_rescale,
output_type='latent' if hasattr(shared.sd_model, 'vae') else 'np',
@@ -341,7 +345,7 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
model=shared.sd_refiner,
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,
num_inference_steps=int(p.refiner_steps // (1 - p.refiner_start)) if p.refiner_start > 0 and p.refiner_start < 1 else int(p.refiner_steps // p.denoising_strength + 1),
eta=shared.opts.eta_ddim,
strength=p.denoising_strength,
guidance_scale=p.image_cfg_scale if p.image_cfg_scale is not None else p.cfg_scale,
@@ -356,6 +360,7 @@ def process_diffusers(p: StableDiffusionProcessing, seeds, prompts, negative_pro
)
refiner_output = shared.sd_refiner(**refiner_args) # pylint: disable=not-callable
p.extra_generation_params['Image CFG scale'] = p.image_cfg_scale if p.image_cfg_scale is not None else None
p.extra_generation_params['Refiner steps'] = p.refiner_steps
p.extra_generation_params['Refiner start'] = p.refiner_start
p.extra_generation_params["Hires steps"] = p.hr_second_pass_steps
modules/txt2img.py
+3 -2
View File
@@ -5,9 +5,9 @@ from modules.ui import plaintext_to_html
from modules.memstats import memory_stats
def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_index: int, latent_index: int, full_quality: bool, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, diffusers_guidance_rescale: float, clip_skip: int, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, refiner_start: int, refiner_prompt: str, refiner_negative: str, override_settings_texts, *args): # pylint: disable=unused-argument
def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_index: int, latent_index: int, full_quality: bool, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, diffusers_guidance_rescale: float, clip_skip: int, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, refiner_steps: int, refiner_start: int, refiner_prompt: str, refiner_negative: str, override_settings_texts, *args): # pylint: disable=unused-argument
shared.log.debug(f'txt2img: id_task={id_task}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|steps={steps}|sampler_index={sampler_index}|latent_index={latent_index}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|clip_skip={clip_skip}|seed={seed}|subseed={subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}||height={height}|width={width}|enable_hr={enable_hr}|denoising_strength={denoising_strength}|hr_scale={hr_scale}|hr_upscaler={hr_upscaler}|hr_second_pass_steps={hr_second_pass_steps}|hr_resize_x={hr_resize_x}|hr_resize_y={hr_resize_y}|image_cfg_scale={image_cfg_scale}|diffusers_guidance_rescale={diffusers_guidance_rescale}|refiner_start={refiner_start}||refiner_prompt={refiner_prompt}|refiner_negative={refiner_negative}|override_settings_texts={override_settings_texts}|args={args}')
shared.log.debug(f'txt2img: id_task={id_task}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|steps={steps}|sampler_index={sampler_index}|latent_index={latent_index}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|clip_skip={clip_skip}|seed={seed}|subseed={subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}||height={height}|width={width}|enable_hr={enable_hr}|denoising_strength={denoising_strength}|hr_scale={hr_scale}|hr_upscaler={hr_upscaler}|hr_second_pass_steps={hr_second_pass_steps}|hr_resize_x={hr_resize_x}|hr_resize_y={hr_resize_y}|image_cfg_scale={image_cfg_scale}|diffusers_guidance_rescale={diffusers_guidance_rescale}|refiner_steps={refiner_steps}|refiner_start={refiner_start}||refiner_prompt={refiner_prompt}|refiner_negative={refiner_negative}|override_settings_texts={override_settings_texts}|args={args}')
if shared.sd_model is None:
shared.log.warning('Model not loaded')
@@ -53,6 +53,7 @@ def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, step
hr_second_pass_steps=hr_second_pass_steps,
hr_resize_x=hr_resize_x,
hr_resize_y=hr_resize_y,
refiner_steps=refiner_steps,
refiner_start=refiner_start,
refiner_prompt=refiner_prompt,
refiner_negative=refiner_negative,
modules/ui.py
+8 -2
View File
@@ -395,7 +395,7 @@ def create_ui(startup_timer = None):
hr_second_pass_steps, latent_index = create_sampler_and_steps_selection(modules.sd_samplers.samplers, "txt2img", False)
with FormRow(elem_id="txt2img_hires_fix_row1", variant="compact"):
denoising_strength = gr.Slider(minimum=0.05, maximum=1.0, step=0.01, label='Denoising strength', value=0.3, elem_id="txt2img_denoising_strength")
refiner_start = gr.Slider(minimum=0.0, maximum=1.0, step=0.05, label='Denoise start', value=0.8, elem_id="txt2img_refiner_start")
refiner_steps = gr.Slider(minimum=0, maximum=99, step=1, label="Refiner steps", elem_id="txt2img_refiner_steps", value=4)
with FormRow(elem_id="txt2img_hires_finalres", variant="compact"):
hr_final_resolution = FormHTML(value="", elem_id="txtimg_hr_finalres", label="Upscaled resolution", interactive=False)
with FormRow(elem_id="txt2img_hires_fix_row2", variant="compact"):
@@ -410,6 +410,7 @@ def create_ui(startup_timer = None):
with FormRow(elem_id="txt2img_refiner_row1", variant="compact"):
image_cfg_scale = gr.Slider(minimum=1.1, maximum=30.0, step=0.1, label='Secondary CFG Scale', value=6.0, elem_id="txt2img_image_cfg_scale")
diffusers_guidance_rescale = gr.Slider(minimum=0.0, maximum=1.0, step=0.05, label='Guidance rescale', value=0.7, elem_id="txt2img_image_cfg_rescale")
refiner_start = gr.Slider(minimum=0.0, maximum=1.0, step=0.05, label='Refiner start', value=0.8, elem_id="txt2img_refiner_start")
with FormRow(elem_id="txt2img_refiner_row2", variant="compact"):
refiner_prompt = gr.Textbox(value='', label='Secondary Prompt')
with FormRow(elem_id="txt2img_refiner_row3", variant="compact"):
@@ -453,7 +454,7 @@ def create_ui(startup_timer = None):
height, width,
show_second_pass, denoising_strength,
hr_scale, hr_upscaler, hr_second_pass_steps, hr_resize_x, hr_resize_y,
refiner_start, refiner_prompt, refiner_negative,
refiner_steps, refiner_start, refiner_prompt, refiner_negative,
override_settings,
] + custom_inputs,
outputs=[
@@ -492,6 +493,7 @@ def create_ui(startup_timer = None):
(clip_skip, "Clip skip"),
(latent_index, "Latent sampler"),
(denoising_strength, "Denoising strength"),
(refiner_steps, "Refiner steps"),
(refiner_start, "Refiner start"),
(full_quality, "Full quality"),
(restore_faces, "Face restoration"),
@@ -507,6 +509,7 @@ def create_ui(startup_timer = None):
(hr_resize_y, "Hires resize-2"),
(diffusers_guidance_rescale, "CFG rescale"),
(image_cfg_scale, "Refiner CFG scale"),
(refiner_steps, "Refiner steps"),
(refiner_start, "Refiner start"),
(tiling, "Tiling"),
(refiner_negative, "Negative2"),
@@ -754,6 +757,7 @@ def create_ui(startup_timer = None):
batch_count, batch_size,
cfg_scale, image_cfg_scale,
diffusers_guidance_rescale,
refiner_steps,
refiner_start,
clip_skip,
denoising_strength,
@@ -848,6 +852,7 @@ def create_ui(startup_timer = None):
(clip_skip, "Clip skip"),
(latent_index, "Latent sampler"),
(denoising_strength, "Denoising strength"),
(refiner_steps, "Refiner steps"),
(refiner_start, "Refiner start"),
(full_quality, "Full quality"),
(restore_faces, "Face restoration"),
@@ -863,6 +868,7 @@ def create_ui(startup_timer = None):
(hr_resize_y, "Hires resize-2"),
(diffusers_guidance_rescale, "CFG rescale"),
(image_cfg_scale, "Image CFG scale"),
(refiner_steps, "Refiner steps"),
(refiner_start, "Refiner start"),
(tiling, "Tiling"),
(refiner_negative, "Negative2"),