采样器这块基本都是用的k-diffusion,模型用的是stability的原生项目generative-models中的sgm,这点和fooocus不同,fooocus底层依赖comfyui中的models,comfy是用load_state_dict的方式解析的,用的load_checkpoint_guess_config函数,这个函数webui中也有。
webui在paths中导入了generative-models,在sd_model_config中导入了config.sdxl和config.sdxl_refiner两个config,模型使用sgm下的models/diffusion/DiffusionEngine初始化,refiner和base的模型几乎是一致的。
python webui.py --port 6006 --no-half-vae
python
webui()->
initialize()->
initialize_rest()->
- sd_samplers.py -> set_samplers()->sd_samplers_kdiffusion.py->
- extensions.py -> list_extensions()
- initialize_util.py -> restore_config_state_file()
- sd_models.py -> list_models()
- localization.py -> list_localizations()
- scripts.load_scripts() -> scripts.py
-- scripts_txt2img=ScriptRunner()/scripts_img2img=ScriptRunner()/scripts_postpro=scripts_postprocessing.ScriptPostprocessingRunner()(scripts_postprocessing.py)
- modelloader.py -> load_upscaler()
- sd_vae.py -> refresh_vae_list()
- textual_inversion/textual_inversion.py -> list_textual_inversion_templates()
- script_callbacks.py -> on_list_optimizers(sd_hijack_optimizations.list_optimizers)
- sd_hijack.py -> list_optimizers()
- sd_unet.py -> list_unets()
- load_model -> shared.py
- shared_items.py -> reload_hypernetworks() # 这种方式现在几乎不用了
- ui_extra_networks.py -> initialize()/register_default_pages()
- extra_networks.py -> initialize()/register_default_extra_networks()
ui.py -> ui.create_ui()
一些基础参数的初始化也在这里,关于ui设计在webui中代码还是挺多的
ui_components.py 一些设计的ui组件
shared_items.py 重复的一些item
下面就是一个FormRow:
python
elif category == "dimensions":
with FormRow():
with gr.Column(elem_id="txt2img_column_size", scale=4):
width = gr.Slider(minimum=64, maximum=2048, step=8, label="Width", value=512, elem_id="txt2img_width")
height = gr.Slider(minimum=64, maximum=2048, step=8, label="Height", value=512, elem_id="txt2img_height")
....
调用接口入口:
python
txt2img_args = dict(
fn=wrap_gradio_gpu_call(modules.txt2img.txt2img, extra_outputs=[None, '', '']),
_js="submit",
inputs=[
dummy_component,
toprow.prompt,
toprow.negative_prompt,
toprow.ui_styles.dropdown,
steps,
sampler_name,
batch_count,
batch_size,
cfg_scale,
height,
width,
enable_hr,
denoising_strength,
hr_scale,
hr_upscaler,
hr_second_pass_steps,
hr_resize_x,
hr_resize_y,
hr_checkpoint_name,
hr_sampler_name,
hr_prompt,
hr_negative_prompt,
override_settings,
] + custom_inputs,
python
p = processing.StableDiffusionProcessingTxt2Img(sd_model,,prompt,negative_prompt,sampler_name,...)->
processed = processing.process_images(p)
python
res = process_image_inner(p)
- sample_ddim = p.sample(conditioning,unconditional_conditioning,seeds,subseeds,subseed_strength,prompts)-> StableDiffusionProcessingTxt2Img.sample()
-- self.sampler = sd_sampler.create_sampler(self.sampler_name,self.sd_model)
-- samples = self.samplers.sample(c,uc,image_encoditioning=self.txt2img_image_conditioning(x))
sd_samplers_kdiffusion.py
python
sample()->
samples = self.launch_sampling(steps,lambda:self.func(self.model_wrap_cfg,x,self.sampler_extra_args,...))
model_rap_cfg:CFGDenoiseKDiffusion->sd_samplers_cfg_denoiser.CFGDenoiser
sd_samplers_common.py
python
func() = sample_dpmpp_2m ->
repositories/k-diffusion/k_diffusion/sampling.py
python
sample_dpmpp_2m()->
- denoised = model(x,sigmas[i]*s_in,**extra_args)->
...
# 此处就是去噪产生图片的过程
modules/sd_samplers_cfg_denoiser.py model =
python
CFGDenoiser()->
forward(x:2x4x128x128,sigma:[14.6146,14.6146],uncond:SchedulePromptConditionings,cond:MulticondLearnedConditioning,cond_scale:7,s_min_uncond:0,image_cond:2x5x1x1)->
denoised:2x4x128x128
# unet预测都被封装在这里
modules/sd_models.py 这块主要是
python
reload_model_weights()->
sd_model = reuse_model_from_already_loaded(sd_model,checkpoint_info,...)
load_model()
....
modules/sd_samplers_common.py
python
sd_models.reload_model_weights(refiner_checkpoint_info)
cfg_denoiser.update_inner_model()->
modules/sd_samplers_cfg_denoiser.py
python
forward()->