Merge branch 'master' of github.com:AUTOMATIC1111/stable-diffusion-webui into gamepad
This commit is contained in:
unknown
commit
21766a0898
@ -20,8 +20,7 @@ model:
|
||||
conditioning_key: hybrid
|
||||
monitor: val/loss_simple_ema
|
||||
scale_factor: 0.18215
|
||||
use_ema: true
|
||||
load_ema: true
|
||||
use_ema: false
|
||||
|
||||
scheduler_config: # 10000 warmup steps
|
||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
||||
|
||||
@ -1,4 +1,4 @@
|
||||
from modules import extra_networks
|
||||
from modules import extra_networks, shared
|
||||
import lora
|
||||
|
||||
class ExtraNetworkLora(extra_networks.ExtraNetwork):
|
||||
@ -6,6 +6,12 @@ class ExtraNetworkLora(extra_networks.ExtraNetwork):
|
||||
super().__init__('lora')
|
||||
|
||||
def activate(self, p, params_list):
|
||||
additional = shared.opts.sd_lora
|
||||
|
||||
if additional != "" and additional in lora.available_loras and len([x for x in params_list if x.items[0] == additional]) == 0:
|
||||
p.all_prompts = [x + f"<lora:{additional}:{shared.opts.extra_networks_default_multiplier}>" for x in p.all_prompts]
|
||||
params_list.append(extra_networks.ExtraNetworkParams(items=[additional, shared.opts.extra_networks_default_multiplier]))
|
||||
|
||||
names = []
|
||||
multipliers = []
|
||||
for params in params_list:
|
||||
|
||||
@ -1,4 +1,5 @@
|
||||
import torch
|
||||
import gradio as gr
|
||||
|
||||
import lora
|
||||
import extra_networks_lora
|
||||
@ -31,5 +32,7 @@ script_callbacks.on_before_ui(before_ui)
|
||||
|
||||
|
||||
shared.options_templates.update(shared.options_section(('extra_networks', "Extra Networks"), {
|
||||
"sd_lora": shared.OptionInfo("None", "Add Lora to prompt", gr.Dropdown, lambda: {"choices": [""] + [x for x in lora.available_loras]}, refresh=lora.list_available_loras),
|
||||
"lora_apply_to_outputs": shared.OptionInfo(False, "Apply Lora to outputs rather than inputs when possible (experimental)"),
|
||||
|
||||
}))
|
||||
|
||||
@ -20,13 +20,14 @@ class ExtraNetworksPageLora(ui_extra_networks.ExtraNetworksPage):
|
||||
preview = None
|
||||
for file in previews:
|
||||
if os.path.isfile(file):
|
||||
preview = "./file=" + file.replace('\\', '/') + "?mtime=" + str(os.path.getmtime(file))
|
||||
preview = self.link_preview(file)
|
||||
break
|
||||
|
||||
yield {
|
||||
"name": name,
|
||||
"filename": path,
|
||||
"preview": preview,
|
||||
"search_term": self.search_terms_from_path(lora_on_disk.filename),
|
||||
"prompt": json.dumps(f"<lora:{name}:") + " + opts.extra_networks_default_multiplier + " + json.dumps(">"),
|
||||
"local_preview": path + ".png",
|
||||
}
|
||||
|
||||
@ -4,6 +4,7 @@
|
||||
<ul>
|
||||
<a href="#" title="replace preview image with currently selected in gallery" onclick={save_card_preview}>replace preview</a>
|
||||
</ul>
|
||||
<span style="display:none" class='search_term'>{search_term}</span>
|
||||
</div>
|
||||
<span class='name'>{name}</span>
|
||||
</div>
|
||||
|
||||
@ -1,7 +1,8 @@
|
||||
|
||||
function extensions_apply(_, _){
|
||||
disable = []
|
||||
update = []
|
||||
var disable = []
|
||||
var update = []
|
||||
|
||||
gradioApp().querySelectorAll('#extensions input[type="checkbox"]').forEach(function(x){
|
||||
if(x.name.startsWith("enable_") && ! x.checked)
|
||||
disable.push(x.name.substr(7))
|
||||
@ -16,11 +17,24 @@ function extensions_apply(_, _){
|
||||
}
|
||||
|
||||
function extensions_check(){
|
||||
var disable = []
|
||||
|
||||
gradioApp().querySelectorAll('#extensions input[type="checkbox"]').forEach(function(x){
|
||||
if(x.name.startsWith("enable_") && ! x.checked)
|
||||
disable.push(x.name.substr(7))
|
||||
})
|
||||
|
||||
gradioApp().querySelectorAll('#extensions .extension_status').forEach(function(x){
|
||||
x.innerHTML = "Loading..."
|
||||
})
|
||||
|
||||
return []
|
||||
|
||||
var id = randomId()
|
||||
requestProgress(id, gradioApp().getElementById('extensions_installed_top'), null, function(){
|
||||
|
||||
})
|
||||
|
||||
return [id, JSON.stringify(disable)]
|
||||
}
|
||||
|
||||
function install_extension_from_index(button, url){
|
||||
|
||||
@ -16,7 +16,7 @@ function setupExtraNetworksForTab(tabname){
|
||||
searchTerm = search.value.toLowerCase()
|
||||
|
||||
gradioApp().querySelectorAll('#'+tabname+'_extra_tabs div.card').forEach(function(elem){
|
||||
text = elem.querySelector('.name').textContent.toLowerCase()
|
||||
text = elem.querySelector('.name').textContent.toLowerCase() + " " + elem.querySelector('.search_term').textContent.toLowerCase()
|
||||
elem.style.display = text.indexOf(searchTerm) == -1 ? "none" : ""
|
||||
})
|
||||
});
|
||||
@ -48,10 +48,39 @@ function setupExtraNetworks(){
|
||||
|
||||
onUiLoaded(setupExtraNetworks)
|
||||
|
||||
var re_extranet = /<([^:]+:[^:]+):[\d\.]+>/;
|
||||
var re_extranet_g = /\s+<([^:]+:[^:]+):[\d\.]+>/g;
|
||||
|
||||
function tryToRemoveExtraNetworkFromPrompt(textarea, text){
|
||||
var m = text.match(re_extranet)
|
||||
if(! m) return false
|
||||
|
||||
var partToSearch = m[1]
|
||||
var replaced = false
|
||||
var newTextareaText = textarea.value.replaceAll(re_extranet_g, function(found, index){
|
||||
m = found.match(re_extranet);
|
||||
if(m[1] == partToSearch){
|
||||
replaced = true;
|
||||
return ""
|
||||
}
|
||||
return found;
|
||||
})
|
||||
|
||||
if(replaced){
|
||||
textarea.value = newTextareaText
|
||||
return true;
|
||||
}
|
||||
|
||||
return false
|
||||
}
|
||||
|
||||
function cardClicked(tabname, textToAdd, allowNegativePrompt){
|
||||
var textarea = allowNegativePrompt ? activePromptTextarea[tabname] : gradioApp().querySelector("#" + tabname + "_prompt > label > textarea")
|
||||
|
||||
textarea.value = textarea.value + " " + textToAdd
|
||||
if(! tryToRemoveExtraNetworkFromPrompt(textarea, textToAdd)){
|
||||
textarea.value = textarea.value + " " + textToAdd
|
||||
}
|
||||
|
||||
updateInput(textarea)
|
||||
}
|
||||
|
||||
@ -67,3 +96,12 @@ function saveCardPreview(event, tabname, filename){
|
||||
event.stopPropagation()
|
||||
event.preventDefault()
|
||||
}
|
||||
|
||||
function extraNetworksSearchButton(tabs_id, event){
|
||||
searchTextarea = gradioApp().querySelector("#" + tabs_id + ' > div > textarea')
|
||||
button = event.target
|
||||
text = button.classList.contains("search-all") ? "" : button.textContent.trim()
|
||||
|
||||
searchTextarea.value = text
|
||||
updateInput(searchTextarea)
|
||||
}
|
||||
@ -191,6 +191,28 @@ function confirm_clear_prompt(prompt, negative_prompt) {
|
||||
return [prompt, negative_prompt]
|
||||
}
|
||||
|
||||
|
||||
promptTokecountUpdateFuncs = {}
|
||||
|
||||
function recalculatePromptTokens(name){
|
||||
if(promptTokecountUpdateFuncs[name]){
|
||||
promptTokecountUpdateFuncs[name]()
|
||||
}
|
||||
}
|
||||
|
||||
function recalculate_prompts_txt2img(){
|
||||
recalculatePromptTokens('txt2img_prompt')
|
||||
recalculatePromptTokens('txt2img_neg_prompt')
|
||||
return args_to_array(arguments);
|
||||
}
|
||||
|
||||
function recalculate_prompts_img2img(){
|
||||
recalculatePromptTokens('img2img_prompt')
|
||||
recalculatePromptTokens('img2img_neg_prompt')
|
||||
return args_to_array(arguments);
|
||||
}
|
||||
|
||||
|
||||
opts = {}
|
||||
onUiUpdate(function(){
|
||||
if(Object.keys(opts).length != 0) return;
|
||||
@ -232,14 +254,12 @@ onUiUpdate(function(){
|
||||
return
|
||||
}
|
||||
|
||||
|
||||
prompt.parentElement.insertBefore(counter, prompt)
|
||||
counter.classList.add("token-counter")
|
||||
prompt.parentElement.style.position = "relative"
|
||||
|
||||
textarea.addEventListener("input", function(){
|
||||
update_token_counter(id_button);
|
||||
});
|
||||
promptTokecountUpdateFuncs[id] = function(){ update_token_counter(id_button); }
|
||||
textarea.addEventListener("input", promptTokecountUpdateFuncs[id]);
|
||||
}
|
||||
|
||||
registerTextarea('txt2img_prompt', 'txt2img_token_counter', 'txt2img_token_button')
|
||||
@ -273,7 +293,7 @@ onOptionsChanged(function(){
|
||||
|
||||
let txt2img_textarea, img2img_textarea = undefined;
|
||||
let wait_time = 800
|
||||
let token_timeout;
|
||||
let token_timeouts = {};
|
||||
|
||||
function update_txt2img_tokens(...args) {
|
||||
update_token_counter("txt2img_token_button")
|
||||
@ -290,9 +310,9 @@ function update_img2img_tokens(...args) {
|
||||
}
|
||||
|
||||
function update_token_counter(button_id) {
|
||||
if (token_timeout)
|
||||
clearTimeout(token_timeout);
|
||||
token_timeout = setTimeout(() => gradioApp().getElementById(button_id)?.click(), wait_time);
|
||||
if (token_timeouts[button_id])
|
||||
clearTimeout(token_timeouts[button_id]);
|
||||
token_timeouts[button_id] = setTimeout(() => gradioApp().getElementById(button_id)?.click(), wait_time);
|
||||
}
|
||||
|
||||
function restart_reload(){
|
||||
@ -309,3 +329,10 @@ function updateInput(target){
|
||||
Object.defineProperty(e, "target", {value: target})
|
||||
target.dispatchEvent(e);
|
||||
}
|
||||
|
||||
|
||||
var desiredCheckpointName = null;
|
||||
function selectCheckpoint(name){
|
||||
desiredCheckpointName = name;
|
||||
gradioApp().getElementById('change_checkpoint').click()
|
||||
}
|
||||
|
||||
@ -223,6 +223,7 @@ def prepare_environment():
|
||||
requirements_file = os.environ.get('REQS_FILE', "requirements_versions.txt")
|
||||
commandline_args = os.environ.get('COMMANDLINE_ARGS', "")
|
||||
|
||||
xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.16rc425')
|
||||
gfpgan_package = os.environ.get('GFPGAN_PACKAGE', "git+https://github.com/TencentARC/GFPGAN.git@8d2447a2d918f8eba5a4a01463fd48e45126a379")
|
||||
clip_package = os.environ.get('CLIP_PACKAGE', "git+https://github.com/openai/CLIP.git@d50d76daa670286dd6cacf3bcd80b5e4823fc8e1")
|
||||
openclip_package = os.environ.get('OPENCLIP_PACKAGE', "git+https://github.com/mlfoundations/open_clip.git@bb6e834e9c70d9c27d0dc3ecedeebeaeb1ffad6b")
|
||||
@ -282,7 +283,7 @@ def prepare_environment():
|
||||
if (not is_installed("xformers") or reinstall_xformers) and xformers:
|
||||
if platform.system() == "Windows":
|
||||
if platform.python_version().startswith("3.10"):
|
||||
run_pip(f"install -U -I --no-deps xformers==0.0.16rc425", "xformers")
|
||||
run_pip(f"install -U -I --no-deps {xformers_package}", "xformers")
|
||||
else:
|
||||
print("Installation of xformers is not supported in this version of Python.")
|
||||
print("You can also check this and build manually: https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Xformers#building-xformers-on-windows-by-duckness")
|
||||
|
||||
@ -87,6 +87,14 @@ dtype_unet = torch.float16
|
||||
unet_needs_upcast = False
|
||||
|
||||
|
||||
def cond_cast_unet(input):
|
||||
return input.to(dtype_unet) if unet_needs_upcast else input
|
||||
|
||||
|
||||
def cond_cast_float(input):
|
||||
return input.float() if unet_needs_upcast else input
|
||||
|
||||
|
||||
def randn(seed, shape):
|
||||
torch.manual_seed(seed)
|
||||
if device.type == 'mps':
|
||||
@ -199,6 +207,3 @@ if has_mps():
|
||||
cumsum_needs_bool_fix = not torch.BoolTensor([True,True]).to(device=torch.device("mps"), dtype=torch.int64).equal(torch.BoolTensor([True,False]).to(torch.device("mps")).cumsum(0))
|
||||
torch.cumsum = lambda input, *args, **kwargs: ( cumsum_fix(input, orig_cumsum, *args, **kwargs) )
|
||||
torch.Tensor.cumsum = lambda self, *args, **kwargs: ( cumsum_fix(self, orig_Tensor_cumsum, *args, **kwargs) )
|
||||
orig_narrow = torch.narrow
|
||||
torch.narrow = lambda *args, **kwargs: ( orig_narrow(*args, **kwargs).clone() )
|
||||
|
||||
|
||||
@ -1,4 +1,4 @@
|
||||
from modules import extra_networks
|
||||
from modules import extra_networks, shared, extra_networks
|
||||
from modules.hypernetworks import hypernetwork
|
||||
|
||||
|
||||
@ -7,6 +7,12 @@ class ExtraNetworkHypernet(extra_networks.ExtraNetwork):
|
||||
super().__init__('hypernet')
|
||||
|
||||
def activate(self, p, params_list):
|
||||
additional = shared.opts.sd_hypernetwork
|
||||
|
||||
if additional != "" and additional in shared.hypernetworks and len([x for x in params_list if x.items[0] == additional]) == 0:
|
||||
p.all_prompts = [x + f"<hypernet:{additional}:{shared.opts.extra_networks_default_multiplier}>" for x in p.all_prompts]
|
||||
params_list.append(extra_networks.ExtraNetworkParams(items=[additional, shared.opts.extra_networks_default_multiplier]))
|
||||
|
||||
names = []
|
||||
multipliers = []
|
||||
for params in params_list:
|
||||
|
||||
@ -1,4 +1,5 @@
|
||||
import base64
|
||||
import html
|
||||
import io
|
||||
import math
|
||||
import os
|
||||
@ -11,19 +12,28 @@ from modules import shared, ui_tempdir, script_callbacks
|
||||
import tempfile
|
||||
from PIL import Image
|
||||
|
||||
re_param_code = r'\s*([\w ]+):\s*("(?:\\|\"|[^\"])+"|[^,]*)(?:,|$)'
|
||||
re_param_code = r'\s*([\w ]+):\s*("(?:\\"[^,]|\\"|\\|[^\"])+"|[^,]*)(?:,|$)'
|
||||
re_param = re.compile(re_param_code)
|
||||
re_params = re.compile(r"^(?:" + re_param_code + "){3,}$")
|
||||
re_imagesize = re.compile(r"^(\d+)x(\d+)$")
|
||||
re_hypernet_hash = re.compile("\(([0-9a-f]+)\)$")
|
||||
type_of_gr_update = type(gr.update())
|
||||
|
||||
paste_fields = {}
|
||||
bind_list = []
|
||||
registered_param_bindings = []
|
||||
|
||||
|
||||
class ParamBinding:
|
||||
def __init__(self, paste_button, tabname, source_text_component=None, source_image_component=None, source_tabname=None, override_settings_component=None):
|
||||
self.paste_button = paste_button
|
||||
self.tabname = tabname
|
||||
self.source_text_component = source_text_component
|
||||
self.source_image_component = source_image_component
|
||||
self.source_tabname = source_tabname
|
||||
self.override_settings_component = override_settings_component
|
||||
|
||||
|
||||
def reset():
|
||||
paste_fields.clear()
|
||||
bind_list.clear()
|
||||
|
||||
|
||||
def quote(text):
|
||||
@ -75,26 +85,6 @@ def add_paste_fields(tabname, init_img, fields):
|
||||
modules.ui.img2img_paste_fields = fields
|
||||
|
||||
|
||||
def integrate_settings_paste_fields(component_dict):
|
||||
from modules import ui
|
||||
|
||||
settings_map = {
|
||||
'CLIP_stop_at_last_layers': 'Clip skip',
|
||||
'inpainting_mask_weight': 'Conditional mask weight',
|
||||
'sd_model_checkpoint': 'Model hash',
|
||||
'eta_noise_seed_delta': 'ENSD',
|
||||
'initial_noise_multiplier': 'Noise multiplier',
|
||||
}
|
||||
settings_paste_fields = [
|
||||
(component_dict[k], lambda d, k=k, v=v: ui.apply_setting(k, d.get(v, None)))
|
||||
for k, v in settings_map.items()
|
||||
]
|
||||
|
||||
for tabname, info in paste_fields.items():
|
||||
if info["fields"] is not None:
|
||||
info["fields"] += settings_paste_fields
|
||||
|
||||
|
||||
def create_buttons(tabs_list):
|
||||
buttons = {}
|
||||
for tab in tabs_list:
|
||||
@ -102,9 +92,60 @@ def create_buttons(tabs_list):
|
||||
return buttons
|
||||
|
||||
|
||||
#if send_generate_info is a tab name, mean generate_info comes from the params fields of the tab
|
||||
def bind_buttons(buttons, send_image, send_generate_info):
|
||||
bind_list.append([buttons, send_image, send_generate_info])
|
||||
"""old function for backwards compatibility; do not use this, use register_paste_params_button"""
|
||||
for tabname, button in buttons.items():
|
||||
source_text_component = send_generate_info if isinstance(send_generate_info, gr.components.Component) else None
|
||||
source_tabname = send_generate_info if isinstance(send_generate_info, str) else None
|
||||
|
||||
register_paste_params_button(ParamBinding(paste_button=button, tabname=tabname, source_text_component=source_text_component, source_image_component=send_image, source_tabname=source_tabname))
|
||||
|
||||
|
||||
def register_paste_params_button(binding: ParamBinding):
|
||||
registered_param_bindings.append(binding)
|
||||
|
||||
|
||||
def connect_paste_params_buttons():
|
||||
binding: ParamBinding
|
||||
for binding in registered_param_bindings:
|
||||
destination_image_component = paste_fields[binding.tabname]["init_img"]
|
||||
fields = paste_fields[binding.tabname]["fields"]
|
||||
|
||||
destination_width_component = next(iter([field for field, name in fields if name == "Size-1"] if fields else []), None)
|
||||
destination_height_component = next(iter([field for field, name in fields if name == "Size-2"] if fields else []), None)
|
||||
|
||||
if binding.source_image_component and destination_image_component:
|
||||
if isinstance(binding.source_image_component, gr.Gallery):
|
||||
func = send_image_and_dimensions if destination_width_component else image_from_url_text
|
||||
jsfunc = "extract_image_from_gallery"
|
||||
else:
|
||||
func = send_image_and_dimensions if destination_width_component else lambda x: x
|
||||
jsfunc = None
|
||||
|
||||
binding.paste_button.click(
|
||||
fn=func,
|
||||
_js=jsfunc,
|
||||
inputs=[binding.source_image_component],
|
||||
outputs=[destination_image_component, destination_width_component, destination_height_component] if destination_width_component else [destination_image_component],
|
||||
)
|
||||
|
||||
if binding.source_text_component is not None and fields is not None:
|
||||
connect_paste(binding.paste_button, fields, binding.source_text_component, binding.override_settings_component, binding.tabname)
|
||||
|
||||
if binding.source_tabname is not None and fields is not None:
|
||||
paste_field_names = ['Prompt', 'Negative prompt', 'Steps', 'Face restoration'] + (["Seed"] if shared.opts.send_seed else [])
|
||||
binding.paste_button.click(
|
||||
fn=lambda *x: x,
|
||||
inputs=[field for field, name in paste_fields[binding.source_tabname]["fields"] if name in paste_field_names],
|
||||
outputs=[field for field, name in fields if name in paste_field_names],
|
||||
)
|
||||
|
||||
binding.paste_button.click(
|
||||
fn=None,
|
||||
_js=f"switch_to_{binding.tabname}",
|
||||
inputs=None,
|
||||
outputs=None,
|
||||
)
|
||||
|
||||
|
||||
def send_image_and_dimensions(x):
|
||||
@ -123,49 +164,6 @@ def send_image_and_dimensions(x):
|
||||
return img, w, h
|
||||
|
||||
|
||||
def run_bind():
|
||||
for buttons, source_image_component, send_generate_info in bind_list:
|
||||
for tab in buttons:
|
||||
button = buttons[tab]
|
||||
destination_image_component = paste_fields[tab]["init_img"]
|
||||
fields = paste_fields[tab]["fields"]
|
||||
|
||||
destination_width_component = next(iter([field for field, name in fields if name == "Size-1"] if fields else []), None)
|
||||
destination_height_component = next(iter([field for field, name in fields if name == "Size-2"] if fields else []), None)
|
||||
|
||||
if source_image_component and destination_image_component:
|
||||
if isinstance(source_image_component, gr.Gallery):
|
||||
func = send_image_and_dimensions if destination_width_component else image_from_url_text
|
||||
jsfunc = "extract_image_from_gallery"
|
||||
else:
|
||||
func = send_image_and_dimensions if destination_width_component else lambda x: x
|
||||
jsfunc = None
|
||||
|
||||
button.click(
|
||||
fn=func,
|
||||
_js=jsfunc,
|
||||
inputs=[source_image_component],
|
||||
outputs=[destination_image_component, destination_width_component, destination_height_component] if destination_width_component else [destination_image_component],
|
||||
)
|
||||
|
||||
if send_generate_info and fields is not None:
|
||||
if send_generate_info in paste_fields:
|
||||
paste_field_names = ['Prompt', 'Negative prompt', 'Steps', 'Face restoration'] + (["Seed"] if shared.opts.send_seed else [])
|
||||
button.click(
|
||||
fn=lambda *x: x,
|
||||
inputs=[field for field, name in paste_fields[send_generate_info]["fields"] if name in paste_field_names],
|
||||
outputs=[field for field, name in fields if name in paste_field_names],
|
||||
)
|
||||
else:
|
||||
connect_paste(button, fields, send_generate_info)
|
||||
|
||||
button.click(
|
||||
fn=None,
|
||||
_js=f"switch_to_{tab}",
|
||||
inputs=None,
|
||||
outputs=None,
|
||||
)
|
||||
|
||||
|
||||
def find_hypernetwork_key(hypernet_name, hypernet_hash=None):
|
||||
"""Determines the config parameter name to use for the hypernet based on the parameters in the infotext.
|
||||
@ -243,7 +241,7 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model
|
||||
done_with_prompt = False
|
||||
|
||||
*lines, lastline = x.strip().split("\n")
|
||||
if not re_params.match(lastline):
|
||||
if len(re_param.findall(lastline)) < 3:
|
||||
lines.append(lastline)
|
||||
lastline = ''
|
||||
|
||||
@ -262,6 +260,7 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model
|
||||
res["Negative prompt"] = negative_prompt
|
||||
|
||||
for k, v in re_param.findall(lastline):
|
||||
v = v[1:-1] if v[0] == '"' and v[-1] == '"' else v
|
||||
m = re_imagesize.match(v)
|
||||
if m is not None:
|
||||
res[k+"-1"] = m.group(1)
|
||||
@ -286,7 +285,50 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model
|
||||
return res
|
||||
|
||||
|
||||
def connect_paste(button, paste_fields, input_comp, jsfunc=None):
|
||||
settings_map = {}
|
||||
|
||||
infotext_to_setting_name_mapping = [
|
||||
('Clip skip', 'CLIP_stop_at_last_layers', ),
|
||||
('Conditional mask weight', 'inpainting_mask_weight'),
|
||||
('Model hash', 'sd_model_checkpoint'),
|
||||
('ENSD', 'eta_noise_seed_delta'),
|
||||
('Noise multiplier', 'initial_noise_multiplier'),
|
||||
('Eta', 'eta_ancestral'),
|
||||
('Eta DDIM', 'eta_ddim'),
|
||||
('Discard penultimate sigma', 'always_discard_next_to_last_sigma')
|
||||
]
|
||||
|
||||
|
||||
def create_override_settings_dict(text_pairs):
|
||||
"""creates processing's override_settings parameters from gradio's multiselect
|
||||
|
||||
Example input:
|
||||
['Clip skip: 2', 'Model hash: e6e99610c4', 'ENSD: 31337']
|
||||
|
||||
Example output:
|
||||
{'CLIP_stop_at_last_layers': 2, 'sd_model_checkpoint': 'e6e99610c4', 'eta_noise_seed_delta': 31337}
|
||||
"""
|
||||
|
||||
res = {}
|
||||
|
||||
params = {}
|
||||
for pair in text_pairs:
|
||||
k, v = pair.split(":", maxsplit=1)
|
||||
|
||||
params[k] = v.strip()
|
||||
|
||||
for param_name, setting_name in infotext_to_setting_name_mapping:
|
||||
value = params.get(param_name, None)
|
||||
|
||||
if value is None:
|
||||
continue
|
||||
|
||||
res[setting_name] = shared.opts.cast_value(setting_name, value)
|
||||
|
||||
return res
|
||||
|
||||
|
||||
def connect_paste(button, paste_fields, input_comp, override_settings_component, tabname):
|
||||
def paste_func(prompt):
|
||||
if not prompt and not shared.cmd_opts.hide_ui_dir_config:
|
||||
filename = os.path.join(data_path, "params.txt")
|
||||
@ -323,9 +365,35 @@ def connect_paste(button, paste_fields, input_comp, jsfunc=None):
|
||||
|
||||
return res
|
||||
|
||||
if override_settings_component is not None:
|
||||
def paste_settings(params):
|
||||
vals = {}
|
||||
|
||||
for param_name, setting_name in infotext_to_setting_name_mapping:
|
||||
v = params.get(param_name, None)
|
||||
if v is None:
|
||||
continue
|
||||
|
||||
if setting_name == "sd_model_checkpoint" and shared.opts.disable_weights_auto_swap:
|
||||
continue
|
||||
|
||||
v = shared.opts.cast_value(setting_name, v)
|
||||
current_value = getattr(shared.opts, setting_name, None)
|
||||
|
||||
if v == current_value:
|
||||
continue
|
||||
|
||||
vals[param_name] = v
|
||||
|
||||
vals_pairs = [f"{k}: {v}" for k, v in vals.items()]
|
||||
|
||||
return gr.Dropdown.update(value=vals_pairs, choices=vals_pairs, visible=len(vals_pairs) > 0)
|
||||
|
||||
paste_fields = paste_fields + [(override_settings_component, paste_settings)]
|
||||
|
||||
button.click(
|
||||
fn=paste_func,
|
||||
_js=jsfunc,
|
||||
_js=f"recalculate_prompts_{tabname}",
|
||||
inputs=[input_comp],
|
||||
outputs=[x[0] for x in paste_fields],
|
||||
)
|
||||
|
||||
@ -36,6 +36,8 @@ def image_grid(imgs, batch_size=1, rows=None):
|
||||
else:
|
||||
rows = math.sqrt(len(imgs))
|
||||
rows = round(rows)
|
||||
if rows > len(imgs):
|
||||
rows = len(imgs)
|
||||
|
||||
cols = math.ceil(len(imgs) / rows)
|
||||
|
||||
|
||||
@ -7,6 +7,7 @@ import numpy as np
|
||||
from PIL import Image, ImageOps, ImageFilter, ImageEnhance, ImageChops
|
||||
|
||||
from modules import devices, sd_samplers
|
||||
from modules.generation_parameters_copypaste import create_override_settings_dict
|
||||
from modules.processing import Processed, StableDiffusionProcessingImg2Img, process_images
|
||||
from modules.shared import opts, state
|
||||
import modules.shared as shared
|
||||
@ -21,8 +22,10 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args):
|
||||
|
||||
images = shared.listfiles(input_dir)
|
||||
|
||||
inpaint_masks = shared.listfiles(inpaint_mask_dir)
|
||||
is_inpaint_batch = inpaint_mask_dir and len(inpaint_masks) > 0
|
||||
is_inpaint_batch = False
|
||||
if inpaint_mask_dir:
|
||||
inpaint_masks = shared.listfiles(inpaint_mask_dir)
|
||||
is_inpaint_batch = len(inpaint_masks) > 0
|
||||
if is_inpaint_batch:
|
||||
print(f"\nInpaint batch is enabled. {len(inpaint_masks)} masks found.")
|
||||
|
||||
@ -73,7 +76,9 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args):
|
||||
processed_image.save(os.path.join(output_dir, filename))
|
||||
|
||||
|
||||
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, mask_blur: int, mask_alpha: float, inpainting_fill: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, *args):
|
||||
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, mask_blur: int, mask_alpha: float, inpainting_fill: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, denoising_strength: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, height: int, width: int, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, *args):
|
||||
override_settings = create_override_settings_dict(override_settings_texts)
|
||||
|
||||
is_batch = mode == 5
|
||||
|
||||
if mode == 0: # img2img
|
||||
@ -140,6 +145,7 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s
|
||||
inpaint_full_res=inpaint_full_res,
|
||||
inpaint_full_res_padding=inpaint_full_res_padding,
|
||||
inpainting_mask_invert=inpainting_mask_invert,
|
||||
override_settings=override_settings,
|
||||
)
|
||||
|
||||
p.scripts = modules.scripts.scripts_txt2img
|
||||
|
||||
@ -173,8 +173,7 @@ class StableDiffusionProcessing:
|
||||
midas_in = torch.from_numpy(transformed["midas_in"][None, ...]).to(device=shared.device)
|
||||
midas_in = repeat(midas_in, "1 ... -> n ...", n=self.batch_size)
|
||||
|
||||
conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(source_image.to(devices.dtype_vae) if devices.unet_needs_upcast else source_image))
|
||||
conditioning_image = conditioning_image.float() if devices.unet_needs_upcast else conditioning_image
|
||||
conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(source_image))
|
||||
conditioning = torch.nn.functional.interpolate(
|
||||
self.sd_model.depth_model(midas_in),
|
||||
size=conditioning_image.shape[2:],
|
||||
@ -218,7 +217,7 @@ class StableDiffusionProcessing:
|
||||
)
|
||||
|
||||
# Encode the new masked image using first stage of network.
|
||||
conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(conditioning_image.to(devices.dtype_vae) if devices.unet_needs_upcast else conditioning_image))
|
||||
conditioning_image = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(conditioning_image))
|
||||
|
||||
# Create the concatenated conditioning tensor to be fed to `c_concat`
|
||||
conditioning_mask = torch.nn.functional.interpolate(conditioning_mask, size=latent_image.shape[-2:])
|
||||
@ -229,16 +228,18 @@ class StableDiffusionProcessing:
|
||||
return image_conditioning
|
||||
|
||||
def img2img_image_conditioning(self, source_image, latent_image, image_mask=None):
|
||||
source_image = devices.cond_cast_float(source_image)
|
||||
|
||||
# HACK: Using introspection as the Depth2Image model doesn't appear to uniquely
|
||||
# identify itself with a field common to all models. The conditioning_key is also hybrid.
|
||||
if isinstance(self.sd_model, LatentDepth2ImageDiffusion):
|
||||
return self.depth2img_image_conditioning(source_image.float() if devices.unet_needs_upcast else source_image)
|
||||
return self.depth2img_image_conditioning(source_image)
|
||||
|
||||
if self.sd_model.cond_stage_key == "edit":
|
||||
return self.edit_image_conditioning(source_image)
|
||||
|
||||
if self.sampler.conditioning_key in {'hybrid', 'concat'}:
|
||||
return self.inpainting_image_conditioning(source_image.float() if devices.unet_needs_upcast else source_image, latent_image, image_mask=image_mask)
|
||||
return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask)
|
||||
|
||||
# Dummy zero conditioning if we're not using inpainting or depth model.
|
||||
return latent_image.new_zeros(latent_image.shape[0], 5, 1, 1)
|
||||
@ -418,7 +419,7 @@ def create_random_tensors(shape, seeds, subseeds=None, subseed_strength=0.0, see
|
||||
|
||||
def decode_first_stage(model, x):
|
||||
with devices.autocast(disable=x.dtype == devices.dtype_vae):
|
||||
x = model.decode_first_stage(x.to(devices.dtype_vae) if devices.unet_needs_upcast else x)
|
||||
x = model.decode_first_stage(x)
|
||||
|
||||
return x
|
||||
|
||||
@ -449,14 +450,11 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter
|
||||
"Size": f"{p.width}x{p.height}",
|
||||
"Model hash": getattr(p, 'sd_model_hash', None if not opts.add_model_hash_to_info or not shared.sd_model.sd_model_hash else shared.sd_model.sd_model_hash),
|
||||
"Model": (None if not opts.add_model_name_to_info or not shared.sd_model.sd_checkpoint_info.model_name else shared.sd_model.sd_checkpoint_info.model_name.replace(',', '').replace(':', '')),
|
||||
"Batch size": (None if p.batch_size < 2 else p.batch_size),
|
||||
"Batch pos": (None if p.batch_size < 2 else position_in_batch),
|
||||
"Variation seed": (None if p.subseed_strength == 0 else all_subseeds[index]),
|
||||
"Variation seed strength": (None if p.subseed_strength == 0 else p.subseed_strength),
|
||||
"Seed resize from": (None if p.seed_resize_from_w == 0 or p.seed_resize_from_h == 0 else f"{p.seed_resize_from_w}x{p.seed_resize_from_h}"),
|
||||
"Denoising strength": getattr(p, 'denoising_strength', None),
|
||||
"Conditional mask weight": getattr(p, "inpainting_mask_weight", shared.opts.inpainting_mask_weight) if p.is_using_inpainting_conditioning else None,
|
||||
"Eta": (None if p.sampler is None or p.sampler.eta == p.sampler.default_eta else p.sampler.eta),
|
||||
"Clip skip": None if clip_skip <= 1 else clip_skip,
|
||||
"ENSD": None if opts.eta_noise_seed_delta == 0 else opts.eta_noise_seed_delta,
|
||||
}
|
||||
@ -1007,7 +1005,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing):
|
||||
|
||||
image = torch.from_numpy(batch_images)
|
||||
image = 2. * image - 1.
|
||||
image = image.to(device=shared.device, dtype=devices.dtype_vae if devices.unet_needs_upcast else None)
|
||||
image = image.to(shared.device)
|
||||
|
||||
self.init_latent = self.sd_model.get_first_stage_encoding(self.sd_model.encode_first_stage(image))
|
||||
|
||||
|
||||
@ -46,7 +46,7 @@ class UpscalerRealESRGAN(Upscaler):
|
||||
scale=info.scale,
|
||||
model_path=info.local_data_path,
|
||||
model=info.model(),
|
||||
half=not cmd_opts.no_half,
|
||||
half=not cmd_opts.no_half and not cmd_opts.upcast_sampling,
|
||||
tile=opts.ESRGAN_tile,
|
||||
tile_pad=opts.ESRGAN_tile_overlap,
|
||||
)
|
||||
|
||||
@ -345,6 +345,20 @@ class ScriptRunner:
|
||||
outputs=[script.group for script in self.selectable_scripts]
|
||||
)
|
||||
|
||||
self.script_load_ctr = 0
|
||||
def onload_script_visibility(params):
|
||||
title = params.get('Script', None)
|
||||
if title:
|
||||
title_index = self.titles.index(title)
|
||||
visibility = title_index == self.script_load_ctr
|
||||
self.script_load_ctr = (self.script_load_ctr + 1) % len(self.titles)
|
||||
return gr.update(visible=visibility)
|
||||
else:
|
||||
return gr.update(visible=False)
|
||||
|
||||
self.infotext_fields.append( (dropdown, lambda x: gr.update(value=x.get('Script', 'None'))) )
|
||||
self.infotext_fields.extend( [(script.group, onload_script_visibility) for script in self.selectable_scripts] )
|
||||
|
||||
return inputs
|
||||
|
||||
def run(self, p, *args):
|
||||
|
||||
@ -131,6 +131,8 @@ class StableDiffusionModelHijack:
|
||||
m.cond_stage_model.wrapped.model.token_embedding = m.cond_stage_model.wrapped.model.token_embedding.wrapped
|
||||
m.cond_stage_model = m.cond_stage_model.wrapped
|
||||
|
||||
undo_optimizations()
|
||||
|
||||
self.apply_circular(False)
|
||||
self.layers = None
|
||||
self.clip = None
|
||||
@ -171,7 +173,7 @@ class EmbeddingsWithFixes(torch.nn.Module):
|
||||
vecs = []
|
||||
for fixes, tensor in zip(batch_fixes, inputs_embeds):
|
||||
for offset, embedding in fixes:
|
||||
emb = embedding.vec
|
||||
emb = devices.cond_cast_unet(embedding.vec)
|
||||
emb_len = min(tensor.shape[0] - offset - 1, emb.shape[0])
|
||||
tensor = torch.cat([tensor[0:offset + 1], emb[0:emb_len], tensor[offset + 1 + emb_len:]])
|
||||
|
||||
|
||||
@ -55,8 +55,14 @@ class GELUHijack(torch.nn.GELU, torch.nn.Module):
|
||||
|
||||
unet_needs_upcast = lambda *args, **kwargs: devices.unet_needs_upcast
|
||||
CondFunc('ldm.models.diffusion.ddpm.LatentDiffusion.apply_model', apply_model, unet_needs_upcast)
|
||||
CondFunc('ldm.modules.diffusionmodules.openaimodel.timestep_embedding', lambda orig_func, *args, **kwargs: orig_func(*args, **kwargs).to(devices.dtype_unet), unet_needs_upcast)
|
||||
CondFunc('ldm.modules.diffusionmodules.openaimodel.timestep_embedding', lambda orig_func, timesteps, *args, **kwargs: orig_func(timesteps, *args, **kwargs).to(torch.float32 if timesteps.dtype == torch.int64 else devices.dtype_unet), unet_needs_upcast)
|
||||
if version.parse(torch.__version__) <= version.parse("1.13.1"):
|
||||
CondFunc('ldm.modules.diffusionmodules.util.GroupNorm32.forward', lambda orig_func, self, *args, **kwargs: orig_func(self.float(), *args, **kwargs), unet_needs_upcast)
|
||||
CondFunc('ldm.modules.attention.GEGLU.forward', lambda orig_func, self, x: orig_func(self.float(), x.float()).to(devices.dtype_unet), unet_needs_upcast)
|
||||
CondFunc('open_clip.transformer.ResidualAttentionBlock.__init__', lambda orig_func, *args, **kwargs: kwargs.update({'act_layer': GELUHijack}) and False or orig_func(*args, **kwargs), lambda _, *args, **kwargs: kwargs.get('act_layer') is None or kwargs['act_layer'] == torch.nn.GELU)
|
||||
|
||||
first_stage_cond = lambda _, self, *args, **kwargs: devices.unet_needs_upcast and self.model.diffusion_model.dtype == torch.float16
|
||||
first_stage_sub = lambda orig_func, self, x, **kwargs: orig_func(self, x.to(devices.dtype_vae), **kwargs)
|
||||
CondFunc('ldm.models.diffusion.ddpm.LatentDiffusion.decode_first_stage', first_stage_sub, first_stage_cond)
|
||||
CondFunc('ldm.models.diffusion.ddpm.LatentDiffusion.encode_first_stage', first_stage_sub, first_stage_cond)
|
||||
CondFunc('ldm.models.diffusion.ddpm.LatentDiffusion.get_first_stage_encoding', lambda orig_func, *args, **kwargs: orig_func(*args, **kwargs).float(), first_stage_cond)
|
||||
|
||||
@ -41,6 +41,7 @@ class CheckpointInfo:
|
||||
name = name[1:]
|
||||
|
||||
self.name = name
|
||||
self.name_for_extra = os.path.splitext(os.path.basename(filename))[0]
|
||||
self.model_name = os.path.splitext(name.replace("/", "_").replace("\\", "_"))[0]
|
||||
self.hash = model_hash(filename)
|
||||
|
||||
@ -231,12 +232,10 @@ def get_checkpoint_state_dict(checkpoint_info: CheckpointInfo, timer):
|
||||
|
||||
|
||||
def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer):
|
||||
title = checkpoint_info.title
|
||||
sd_model_hash = checkpoint_info.calculate_shorthash()
|
||||
timer.record("calculate hash")
|
||||
|
||||
if checkpoint_info.title != title:
|
||||
shared.opts.data["sd_model_checkpoint"] = checkpoint_info.title
|
||||
shared.opts.data["sd_model_checkpoint"] = checkpoint_info.title
|
||||
|
||||
if state_dict is None:
|
||||
state_dict = get_checkpoint_state_dict(checkpoint_info, timer)
|
||||
|
||||
@ -1,7 +1,9 @@
|
||||
import re
|
||||
import os
|
||||
|
||||
from modules import shared, paths
|
||||
import torch
|
||||
|
||||
from modules import shared, paths, sd_disable_initialization
|
||||
|
||||
sd_configs_path = shared.sd_configs_path
|
||||
sd_repo_configs_path = os.path.join(paths.paths['Stable Diffusion'], "configs", "stable-diffusion")
|
||||
@ -16,12 +18,51 @@ config_inpainting = os.path.join(sd_configs_path, "v1-inpainting-inference.yaml"
|
||||
config_instruct_pix2pix = os.path.join(sd_configs_path, "instruct-pix2pix.yaml")
|
||||
config_alt_diffusion = os.path.join(sd_configs_path, "alt-diffusion-inference.yaml")
|
||||
|
||||
re_parametrization_v = re.compile(r'-v\b')
|
||||
|
||||
def is_using_v_parameterization_for_sd2(state_dict):
|
||||
"""
|
||||
Detects whether unet in state_dict is using v-parameterization. Returns True if it is. You're welcome.
|
||||
"""
|
||||
|
||||
import ldm.modules.diffusionmodules.openaimodel
|
||||
from modules import devices
|
||||
|
||||
device = devices.cpu
|
||||
|
||||
with sd_disable_initialization.DisableInitialization():
|
||||
unet = ldm.modules.diffusionmodules.openaimodel.UNetModel(
|
||||
use_checkpoint=True,
|
||||
use_fp16=False,
|
||||
image_size=32,
|
||||
in_channels=4,
|
||||
out_channels=4,
|
||||
model_channels=320,
|
||||
attention_resolutions=[4, 2, 1],
|
||||
num_res_blocks=2,
|
||||
channel_mult=[1, 2, 4, 4],
|
||||
num_head_channels=64,
|
||||
use_spatial_transformer=True,
|
||||
use_linear_in_transformer=True,
|
||||
transformer_depth=1,
|
||||
context_dim=1024,
|
||||
legacy=False
|
||||
)
|
||||
unet.eval()
|
||||
|
||||
with torch.no_grad():
|
||||
unet_sd = {k.replace("model.diffusion_model.", ""): v for k, v in state_dict.items() if "model.diffusion_model." in k}
|
||||
unet.load_state_dict(unet_sd, strict=True)
|
||||
unet.to(device=device, dtype=torch.float)
|
||||
|
||||
test_cond = torch.ones((1, 2, 1024), device=device) * 0.5
|
||||
x_test = torch.ones((1, 4, 8, 8), device=device) * 0.5
|
||||
|
||||
out = (unet(x_test, torch.asarray([999], device=device), context=test_cond) - x_test).mean().item()
|
||||
|
||||
return out < -1
|
||||
|
||||
|
||||
def guess_model_config_from_state_dict(sd, filename):
|
||||
fn = os.path.basename(filename)
|
||||
|
||||
sd2_cond_proj_weight = sd.get('cond_stage_model.model.transformer.resblocks.0.attn.in_proj_weight', None)
|
||||
diffusion_model_input = sd.get('model.diffusion_model.input_blocks.0.0.weight', None)
|
||||
|
||||
@ -31,7 +72,7 @@ def guess_model_config_from_state_dict(sd, filename):
|
||||
if sd2_cond_proj_weight is not None and sd2_cond_proj_weight.shape[1] == 1024:
|
||||
if diffusion_model_input.shape[1] == 9:
|
||||
return config_sd2_inpainting
|
||||
elif re.search(re_parametrization_v, fn):
|
||||
elif is_using_v_parameterization_for_sd2(sd):
|
||||
return config_sd2v
|
||||
else:
|
||||
return config_sd2
|
||||
|
||||
@ -1,53 +1,11 @@
|
||||
from collections import namedtuple, deque
|
||||
import numpy as np
|
||||
from math import floor
|
||||
import torch
|
||||
import tqdm
|
||||
from PIL import Image
|
||||
import inspect
|
||||
import k_diffusion.sampling
|
||||
import torchsde._brownian.brownian_interval
|
||||
import ldm.models.diffusion.ddim
|
||||
import ldm.models.diffusion.plms
|
||||
from modules import prompt_parser, devices, processing, images, sd_vae_approx
|
||||
from modules import sd_samplers_compvis, sd_samplers_kdiffusion, shared
|
||||
|
||||
from modules.shared import opts, cmd_opts, state
|
||||
import modules.shared as shared
|
||||
from modules.script_callbacks import CFGDenoiserParams, cfg_denoiser_callback
|
||||
|
||||
|
||||
SamplerData = namedtuple('SamplerData', ['name', 'constructor', 'aliases', 'options'])
|
||||
|
||||
samplers_k_diffusion = [
|
||||
('Euler a', 'sample_euler_ancestral', ['k_euler_a', 'k_euler_ancestral'], {}),
|
||||
('Euler', 'sample_euler', ['k_euler'], {}),
|
||||
('LMS', 'sample_lms', ['k_lms'], {}),
|
||||
('Heun', 'sample_heun', ['k_heun'], {}),
|
||||
('DPM2', 'sample_dpm_2', ['k_dpm_2'], {'discard_next_to_last_sigma': True}),
|
||||
('DPM2 a', 'sample_dpm_2_ancestral', ['k_dpm_2_a'], {'discard_next_to_last_sigma': True}),
|
||||
('DPM++ 2S a', 'sample_dpmpp_2s_ancestral', ['k_dpmpp_2s_a'], {}),
|
||||
('DPM++ 2M', 'sample_dpmpp_2m', ['k_dpmpp_2m'], {}),
|
||||
('DPM++ SDE', 'sample_dpmpp_sde', ['k_dpmpp_sde'], {}),
|
||||
('DPM fast', 'sample_dpm_fast', ['k_dpm_fast'], {}),
|
||||
('DPM adaptive', 'sample_dpm_adaptive', ['k_dpm_ad'], {}),
|
||||
('LMS Karras', 'sample_lms', ['k_lms_ka'], {'scheduler': 'karras'}),
|
||||
('DPM2 Karras', 'sample_dpm_2', ['k_dpm_2_ka'], {'scheduler': 'karras', 'discard_next_to_last_sigma': True}),
|
||||
('DPM2 a Karras', 'sample_dpm_2_ancestral', ['k_dpm_2_a_ka'], {'scheduler': 'karras', 'discard_next_to_last_sigma': True}),
|
||||
('DPM++ 2S a Karras', 'sample_dpmpp_2s_ancestral', ['k_dpmpp_2s_a_ka'], {'scheduler': 'karras'}),
|
||||
('DPM++ 2M Karras', 'sample_dpmpp_2m', ['k_dpmpp_2m_ka'], {'scheduler': 'karras'}),
|
||||
('DPM++ SDE Karras', 'sample_dpmpp_sde', ['k_dpmpp_sde_ka'], {'scheduler': 'karras'}),
|
||||
]
|
||||
|
||||
samplers_data_k_diffusion = [
|
||||
SamplerData(label, lambda model, funcname=funcname: KDiffusionSampler(funcname, model), aliases, options)
|
||||
for label, funcname, aliases, options in samplers_k_diffusion
|
||||
if hasattr(k_diffusion.sampling, funcname)
|
||||
]
|
||||
# imports for functions that previously were here and are used by other modules
|
||||
from modules.sd_samplers_common import samples_to_image_grid, sample_to_image
|
||||
|
||||
all_samplers = [
|
||||
*samplers_data_k_diffusion,
|
||||
SamplerData('DDIM', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.ddim.DDIMSampler, model), [], {}),
|
||||
SamplerData('PLMS', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.plms.PLMSSampler, model), [], {}),
|
||||
*sd_samplers_kdiffusion.samplers_data_k_diffusion,
|
||||
*sd_samplers_compvis.samplers_data_compvis,
|
||||
]
|
||||
all_samplers_map = {x.name: x for x in all_samplers}
|
||||
|
||||
@ -73,8 +31,8 @@ def create_sampler(name, model):
|
||||
def set_samplers():
|
||||
global samplers, samplers_for_img2img
|
||||
|
||||
hidden = set(opts.hide_samplers)
|
||||
hidden_img2img = set(opts.hide_samplers + ['PLMS'])
|
||||
hidden = set(shared.opts.hide_samplers)
|
||||
hidden_img2img = set(shared.opts.hide_samplers + ['PLMS'])
|
||||
|
||||
samplers = [x for x in all_samplers if x.name not in hidden]
|
||||
samplers_for_img2img = [x for x in all_samplers if x.name not in hidden_img2img]
|
||||
@ -87,466 +45,3 @@ def set_samplers():
|
||||
|
||||
|
||||
set_samplers()
|
||||
|
||||
sampler_extra_params = {
|
||||
'sample_euler': ['s_churn', 's_tmin', 's_tmax', 's_noise'],
|
||||
'sample_heun': ['s_churn', 's_tmin', 's_tmax', 's_noise'],
|
||||
'sample_dpm_2': ['s_churn', 's_tmin', 's_tmax', 's_noise'],
|
||||
}
|
||||
|
||||
|
||||
def setup_img2img_steps(p, steps=None):
|
||||
if opts.img2img_fix_steps or steps is not None:
|
||||
requested_steps = (steps or p.steps)
|
||||
steps = int(requested_steps / min(p.denoising_strength, 0.999)) if p.denoising_strength > 0 else 0
|
||||
t_enc = requested_steps - 1
|
||||
else:
|
||||
steps = p.steps
|
||||
t_enc = int(min(p.denoising_strength, 0.999) * steps)
|
||||
|
||||
return steps, t_enc
|
||||
|
||||
|
||||
approximation_indexes = {"Full": 0, "Approx NN": 1, "Approx cheap": 2}
|
||||
|
||||
|
||||
def single_sample_to_image(sample, approximation=None):
|
||||
if approximation is None:
|
||||
approximation = approximation_indexes.get(opts.show_progress_type, 0)
|
||||
|
||||
if approximation == 2:
|
||||
x_sample = sd_vae_approx.cheap_approximation(sample)
|
||||
elif approximation == 1:
|
||||
x_sample = sd_vae_approx.model()(sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach()
|
||||
else:
|
||||
x_sample = processing.decode_first_stage(shared.sd_model, sample.unsqueeze(0))[0]
|
||||
|
||||
x_sample = torch.clamp((x_sample + 1.0) / 2.0, min=0.0, max=1.0)
|
||||
x_sample = 255. * np.moveaxis(x_sample.cpu().numpy(), 0, 2)
|
||||
x_sample = x_sample.astype(np.uint8)
|
||||
return Image.fromarray(x_sample)
|
||||
|
||||
|
||||
def sample_to_image(samples, index=0, approximation=None):
|
||||
return single_sample_to_image(samples[index], approximation)
|
||||
|
||||
|
||||
def samples_to_image_grid(samples, approximation=None):
|
||||
return images.image_grid([single_sample_to_image(sample, approximation) for sample in samples])
|
||||
|
||||
|
||||
def store_latent(decoded):
|
||||
state.current_latent = decoded
|
||||
|
||||
if opts.live_previews_enable and opts.show_progress_every_n_steps > 0 and shared.state.sampling_step % opts.show_progress_every_n_steps == 0:
|
||||
if not shared.parallel_processing_allowed:
|
||||
shared.state.assign_current_image(sample_to_image(decoded))
|
||||
|
||||
|
||||
class InterruptedException(BaseException):
|
||||
pass
|
||||
|
||||
|
||||
class VanillaStableDiffusionSampler:
|
||||
def __init__(self, constructor, sd_model):
|
||||
self.sampler = constructor(sd_model)
|
||||
self.is_plms = hasattr(self.sampler, 'p_sample_plms')
|
||||
self.orig_p_sample_ddim = self.sampler.p_sample_plms if self.is_plms else self.sampler.p_sample_ddim
|
||||
self.mask = None
|
||||
self.nmask = None
|
||||
self.init_latent = None
|
||||
self.sampler_noises = None
|
||||
self.step = 0
|
||||
self.stop_at = None
|
||||
self.eta = None
|
||||
self.default_eta = 0.0
|
||||
self.config = None
|
||||
self.last_latent = None
|
||||
|
||||
self.conditioning_key = sd_model.model.conditioning_key
|
||||
|
||||
def number_of_needed_noises(self, p):
|
||||
return 0
|
||||
|
||||
def launch_sampling(self, steps, func):
|
||||
state.sampling_steps = steps
|
||||
state.sampling_step = 0
|
||||
|
||||
try:
|
||||
return func()
|
||||
except InterruptedException:
|
||||
return self.last_latent
|
||||
|
||||
def p_sample_ddim_hook(self, x_dec, cond, ts, unconditional_conditioning, *args, **kwargs):
|
||||
if state.interrupted or state.skipped:
|
||||
raise InterruptedException
|
||||
|
||||
if self.stop_at is not None and self.step > self.stop_at:
|
||||
raise InterruptedException
|
||||
|
||||
# Have to unwrap the inpainting conditioning here to perform pre-processing
|
||||
image_conditioning = None
|
||||
if isinstance(cond, dict):
|
||||
image_conditioning = cond["c_concat"][0]
|
||||
cond = cond["c_crossattn"][0]
|
||||
unconditional_conditioning = unconditional_conditioning["c_crossattn"][0]
|
||||
|
||||
conds_list, tensor = prompt_parser.reconstruct_multicond_batch(cond, self.step)
|
||||
unconditional_conditioning = prompt_parser.reconstruct_cond_batch(unconditional_conditioning, self.step)
|
||||
|
||||
assert all([len(conds) == 1 for conds in conds_list]), 'composition via AND is not supported for DDIM/PLMS samplers'
|
||||
cond = tensor
|
||||
|
||||
# for DDIM, shapes must match, we can't just process cond and uncond independently;
|
||||
# filling unconditional_conditioning with repeats of the last vector to match length is
|
||||
# not 100% correct but should work well enough
|
||||
if unconditional_conditioning.shape[1] < cond.shape[1]:
|
||||
last_vector = unconditional_conditioning[:, -1:]
|
||||
last_vector_repeated = last_vector.repeat([1, cond.shape[1] - unconditional_conditioning.shape[1], 1])
|
||||
unconditional_conditioning = torch.hstack([unconditional_conditioning, last_vector_repeated])
|
||||
elif unconditional_conditioning.shape[1] > cond.shape[1]:
|
||||
unconditional_conditioning = unconditional_conditioning[:, :cond.shape[1]]
|
||||
|
||||
if self.mask is not None:
|
||||
img_orig = self.sampler.model.q_sample(self.init_latent, ts)
|
||||
x_dec = img_orig * self.mask + self.nmask * x_dec
|
||||
|
||||
# Wrap the image conditioning back up since the DDIM code can accept the dict directly.
|
||||
# Note that they need to be lists because it just concatenates them later.
|
||||
if image_conditioning is not None:
|
||||
cond = {"c_concat": [image_conditioning], "c_crossattn": [cond]}
|
||||
unconditional_conditioning = {"c_concat": [image_conditioning], "c_crossattn": [unconditional_conditioning]}
|
||||
|
||||
res = self.orig_p_sample_ddim(x_dec, cond, ts, unconditional_conditioning=unconditional_conditioning, *args, **kwargs)
|
||||
|
||||
if self.mask is not None:
|
||||
self.last_latent = self.init_latent * self.mask + self.nmask * res[1]
|
||||
else:
|
||||
self.last_latent = res[1]
|
||||
|
||||
store_latent(self.last_latent)
|
||||
|
||||
self.step += 1
|
||||
state.sampling_step = self.step
|
||||
shared.total_tqdm.update()
|
||||
|
||||
return res
|
||||
|
||||
def initialize(self, p):
|
||||
self.eta = p.eta if p.eta is not None else opts.eta_ddim
|
||||
|
||||
for fieldname in ['p_sample_ddim', 'p_sample_plms']:
|
||||
if hasattr(self.sampler, fieldname):
|
||||
setattr(self.sampler, fieldname, self.p_sample_ddim_hook)
|
||||
|
||||
self.mask = p.mask if hasattr(p, 'mask') else None
|
||||
self.nmask = p.nmask if hasattr(p, 'nmask') else None
|
||||
|
||||
def adjust_steps_if_invalid(self, p, num_steps):
|
||||
if (self.config.name == 'DDIM' and p.ddim_discretize == 'uniform') or (self.config.name == 'PLMS'):
|
||||
valid_step = 999 / (1000 // num_steps)
|
||||
if valid_step == floor(valid_step):
|
||||
return int(valid_step) + 1
|
||||
|
||||
return num_steps
|
||||
|
||||
def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None, image_conditioning=None):
|
||||
steps, t_enc = setup_img2img_steps(p, steps)
|
||||
steps = self.adjust_steps_if_invalid(p, steps)
|
||||
self.initialize(p)
|
||||
|
||||
self.sampler.make_schedule(ddim_num_steps=steps, ddim_eta=self.eta, ddim_discretize=p.ddim_discretize, verbose=False)
|
||||
x1 = self.sampler.stochastic_encode(x, torch.tensor([t_enc] * int(x.shape[0])).to(shared.device), noise=noise)
|
||||
|
||||
self.init_latent = x
|
||||
self.last_latent = x
|
||||
self.step = 0
|
||||
|
||||
# Wrap the conditioning models with additional image conditioning for inpainting model
|
||||
if image_conditioning is not None:
|
||||
conditioning = {"c_concat": [image_conditioning], "c_crossattn": [conditioning]}
|
||||
unconditional_conditioning = {"c_concat": [image_conditioning], "c_crossattn": [unconditional_conditioning]}
|
||||
|
||||
samples = self.launch_sampling(t_enc + 1, lambda: self.sampler.decode(x1, conditioning, t_enc, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning))
|
||||
|
||||
return samples
|
||||
|
||||
def sample(self, p, x, conditioning, unconditional_conditioning, steps=None, image_conditioning=None):
|
||||
self.initialize(p)
|
||||
|
||||
self.init_latent = None
|
||||
self.last_latent = x
|
||||
self.step = 0
|
||||
|
||||
steps = self.adjust_steps_if_invalid(p, steps or p.steps)
|
||||
|
||||
# Wrap the conditioning models with additional image conditioning for inpainting model
|
||||
# dummy_for_plms is needed because PLMS code checks the first item in the dict to have the right shape
|
||||
if image_conditioning is not None:
|
||||
conditioning = {"dummy_for_plms": np.zeros((conditioning.shape[0],)), "c_crossattn": [conditioning], "c_concat": [image_conditioning]}
|
||||
unconditional_conditioning = {"c_crossattn": [unconditional_conditioning], "c_concat": [image_conditioning]}
|
||||
|
||||
samples_ddim = self.launch_sampling(steps, lambda: self.sampler.sample(S=steps, conditioning=conditioning, batch_size=int(x.shape[0]), shape=x[0].shape, verbose=False, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning, x_T=x, eta=self.eta)[0])
|
||||
|
||||
return samples_ddim
|
||||
|
||||
|
||||
class CFGDenoiser(torch.nn.Module):
|
||||
def __init__(self, model):
|
||||
super().__init__()
|
||||
self.inner_model = model
|
||||
self.mask = None
|
||||
self.nmask = None
|
||||
self.init_latent = None
|
||||
self.step = 0
|
||||
|
||||
def combine_denoised(self, x_out, conds_list, uncond, cond_scale):
|
||||
denoised_uncond = x_out[-uncond.shape[0]:]
|
||||
denoised = torch.clone(denoised_uncond)
|
||||
|
||||
for i, conds in enumerate(conds_list):
|
||||
for cond_index, weight in conds:
|
||||
denoised[i] += (x_out[cond_index] - denoised_uncond[i]) * (weight * cond_scale)
|
||||
|
||||
return denoised
|
||||
|
||||
def forward(self, x, sigma, uncond, cond, cond_scale, image_cond):
|
||||
if state.interrupted or state.skipped:
|
||||
raise InterruptedException
|
||||
|
||||
conds_list, tensor = prompt_parser.reconstruct_multicond_batch(cond, self.step)
|
||||
uncond = prompt_parser.reconstruct_cond_batch(uncond, self.step)
|
||||
|
||||
batch_size = len(conds_list)
|
||||
repeats = [len(conds_list[i]) for i in range(batch_size)]
|
||||
|
||||
x_in = torch.cat([torch.stack([x[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [x])
|
||||
image_cond_in = torch.cat([torch.stack([image_cond[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [image_cond])
|
||||
sigma_in = torch.cat([torch.stack([sigma[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [sigma])
|
||||
|
||||
denoiser_params = CFGDenoiserParams(x_in, image_cond_in, sigma_in, state.sampling_step, state.sampling_steps)
|
||||
cfg_denoiser_callback(denoiser_params)
|
||||
x_in = denoiser_params.x
|
||||
image_cond_in = denoiser_params.image_cond
|
||||
sigma_in = denoiser_params.sigma
|
||||
|
||||
if tensor.shape[1] == uncond.shape[1]:
|
||||
cond_in = torch.cat([tensor, uncond])
|
||||
|
||||
if shared.batch_cond_uncond:
|
||||
x_out = self.inner_model(x_in, sigma_in, cond={"c_crossattn": [cond_in], "c_concat": [image_cond_in]})
|
||||
else:
|
||||
x_out = torch.zeros_like(x_in)
|
||||
for batch_offset in range(0, x_out.shape[0], batch_size):
|
||||
a = batch_offset
|
||||
b = a + batch_size
|
||||
x_out[a:b] = self.inner_model(x_in[a:b], sigma_in[a:b], cond={"c_crossattn": [cond_in[a:b]], "c_concat": [image_cond_in[a:b]]})
|
||||
else:
|
||||
x_out = torch.zeros_like(x_in)
|
||||
batch_size = batch_size*2 if shared.batch_cond_uncond else batch_size
|
||||
for batch_offset in range(0, tensor.shape[0], batch_size):
|
||||
a = batch_offset
|
||||
b = min(a + batch_size, tensor.shape[0])
|
||||
x_out[a:b] = self.inner_model(x_in[a:b], sigma_in[a:b], cond={"c_crossattn": [tensor[a:b]], "c_concat": [image_cond_in[a:b]]})
|
||||
|
||||
x_out[-uncond.shape[0]:] = self.inner_model(x_in[-uncond.shape[0]:], sigma_in[-uncond.shape[0]:], cond={"c_crossattn": [uncond], "c_concat": [image_cond_in[-uncond.shape[0]:]]})
|
||||
|
||||
devices.test_for_nans(x_out, "unet")
|
||||
|
||||
if opts.live_preview_content == "Prompt":
|
||||
store_latent(x_out[0:uncond.shape[0]])
|
||||
elif opts.live_preview_content == "Negative prompt":
|
||||
store_latent(x_out[-uncond.shape[0]:])
|
||||
|
||||
denoised = self.combine_denoised(x_out, conds_list, uncond, cond_scale)
|
||||
|
||||
if self.mask is not None:
|
||||
denoised = self.init_latent * self.mask + self.nmask * denoised
|
||||
|
||||
self.step += 1
|
||||
|
||||
return denoised
|
||||
|
||||
|
||||
class TorchHijack:
|
||||
def __init__(self, sampler_noises):
|
||||
# Using a deque to efficiently receive the sampler_noises in the same order as the previous index-based
|
||||
# implementation.
|
||||
self.sampler_noises = deque(sampler_noises)
|
||||
|
||||
def __getattr__(self, item):
|
||||
if item == 'randn_like':
|
||||
return self.randn_like
|
||||
|
||||
if hasattr(torch, item):
|
||||
return getattr(torch, item)
|
||||
|
||||
raise AttributeError("'{}' object has no attribute '{}'".format(type(self).__name__, item))
|
||||
|
||||
def randn_like(self, x):
|
||||
if self.sampler_noises:
|
||||
noise = self.sampler_noises.popleft()
|
||||
if noise.shape == x.shape:
|
||||
return noise
|
||||
|
||||
if x.device.type == 'mps':
|
||||
return torch.randn_like(x, device=devices.cpu).to(x.device)
|
||||
else:
|
||||
return torch.randn_like(x)
|
||||
|
||||
|
||||
# MPS fix for randn in torchsde
|
||||
def torchsde_randn(size, dtype, device, seed):
|
||||
if device.type == 'mps':
|
||||
generator = torch.Generator(devices.cpu).manual_seed(int(seed))
|
||||
return torch.randn(size, dtype=dtype, device=devices.cpu, generator=generator).to(device)
|
||||
else:
|
||||
generator = torch.Generator(device).manual_seed(int(seed))
|
||||
return torch.randn(size, dtype=dtype, device=device, generator=generator)
|
||||
|
||||
|
||||
torchsde._brownian.brownian_interval._randn = torchsde_randn
|
||||
|
||||
|
||||
class KDiffusionSampler:
|
||||
def __init__(self, funcname, sd_model):
|
||||
denoiser = k_diffusion.external.CompVisVDenoiser if sd_model.parameterization == "v" else k_diffusion.external.CompVisDenoiser
|
||||
|
||||
self.model_wrap = denoiser(sd_model, quantize=shared.opts.enable_quantization)
|
||||
self.funcname = funcname
|
||||
self.func = getattr(k_diffusion.sampling, self.funcname)
|
||||
self.extra_params = sampler_extra_params.get(funcname, [])
|
||||
self.model_wrap_cfg = CFGDenoiser(self.model_wrap)
|
||||
self.sampler_noises = None
|
||||
self.stop_at = None
|
||||
self.eta = None
|
||||
self.default_eta = 1.0
|
||||
self.config = None
|
||||
self.last_latent = None
|
||||
|
||||
self.conditioning_key = sd_model.model.conditioning_key
|
||||
|
||||
def callback_state(self, d):
|
||||
step = d['i']
|
||||
latent = d["denoised"]
|
||||
if opts.live_preview_content == "Combined":
|
||||
store_latent(latent)
|
||||
self.last_latent = latent
|
||||
|
||||
if self.stop_at is not None and step > self.stop_at:
|
||||
raise InterruptedException
|
||||
|
||||
state.sampling_step = step
|
||||
shared.total_tqdm.update()
|
||||
|
||||
def launch_sampling(self, steps, func):
|
||||
state.sampling_steps = steps
|
||||
state.sampling_step = 0
|
||||
|
||||
try:
|
||||
return func()
|
||||
except InterruptedException:
|
||||
return self.last_latent
|
||||
|
||||
def number_of_needed_noises(self, p):
|
||||
return p.steps
|
||||
|
||||
def initialize(self, p):
|
||||
self.model_wrap_cfg.mask = p.mask if hasattr(p, 'mask') else None
|
||||
self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None
|
||||
self.model_wrap_cfg.step = 0
|
||||
self.eta = p.eta or opts.eta_ancestral
|
||||
|
||||
k_diffusion.sampling.torch = TorchHijack(self.sampler_noises if self.sampler_noises is not None else [])
|
||||
|
||||
extra_params_kwargs = {}
|
||||
for param_name in self.extra_params:
|
||||
if hasattr(p, param_name) and param_name in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs[param_name] = getattr(p, param_name)
|
||||
|
||||
if 'eta' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['eta'] = self.eta
|
||||
|
||||
return extra_params_kwargs
|
||||
|
||||
def get_sigmas(self, p, steps):
|
||||
discard_next_to_last_sigma = self.config is not None and self.config.options.get('discard_next_to_last_sigma', False)
|
||||
if opts.always_discard_next_to_last_sigma and not discard_next_to_last_sigma:
|
||||
discard_next_to_last_sigma = True
|
||||
p.extra_generation_params["Discard penultimate sigma"] = True
|
||||
|
||||
steps += 1 if discard_next_to_last_sigma else 0
|
||||
|
||||
if p.sampler_noise_scheduler_override:
|
||||
sigmas = p.sampler_noise_scheduler_override(steps)
|
||||
elif self.config is not None and self.config.options.get('scheduler', None) == 'karras':
|
||||
sigma_min, sigma_max = (0.1, 10) if opts.use_old_karras_scheduler_sigmas else (self.model_wrap.sigmas[0].item(), self.model_wrap.sigmas[-1].item())
|
||||
|
||||
sigmas = k_diffusion.sampling.get_sigmas_karras(n=steps, sigma_min=sigma_min, sigma_max=sigma_max, device=shared.device)
|
||||
else:
|
||||
sigmas = self.model_wrap.get_sigmas(steps)
|
||||
|
||||
if discard_next_to_last_sigma:
|
||||
sigmas = torch.cat([sigmas[:-2], sigmas[-1:]])
|
||||
|
||||
return sigmas
|
||||
|
||||
def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None, image_conditioning=None):
|
||||
steps, t_enc = setup_img2img_steps(p, steps)
|
||||
|
||||
sigmas = self.get_sigmas(p, steps)
|
||||
|
||||
sigma_sched = sigmas[steps - t_enc - 1:]
|
||||
xi = x + noise * sigma_sched[0]
|
||||
|
||||
extra_params_kwargs = self.initialize(p)
|
||||
if 'sigma_min' in inspect.signature(self.func).parameters:
|
||||
## last sigma is zero which isn't allowed by DPM Fast & Adaptive so taking value before last
|
||||
extra_params_kwargs['sigma_min'] = sigma_sched[-2]
|
||||
if 'sigma_max' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigma_max'] = sigma_sched[0]
|
||||
if 'n' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['n'] = len(sigma_sched) - 1
|
||||
if 'sigma_sched' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigma_sched'] = sigma_sched
|
||||
if 'sigmas' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigmas'] = sigma_sched
|
||||
|
||||
self.model_wrap_cfg.init_latent = x
|
||||
self.last_latent = x
|
||||
|
||||
samples = self.launch_sampling(t_enc + 1, lambda: self.func(self.model_wrap_cfg, xi, extra_args={
|
||||
'cond': conditioning,
|
||||
'image_cond': image_conditioning,
|
||||
'uncond': unconditional_conditioning,
|
||||
'cond_scale': p.cfg_scale
|
||||
}, disable=False, callback=self.callback_state, **extra_params_kwargs))
|
||||
|
||||
return samples
|
||||
|
||||
def sample(self, p, x, conditioning, unconditional_conditioning, steps=None, image_conditioning = None):
|
||||
steps = steps or p.steps
|
||||
|
||||
sigmas = self.get_sigmas(p, steps)
|
||||
|
||||
x = x * sigmas[0]
|
||||
|
||||
extra_params_kwargs = self.initialize(p)
|
||||
if 'sigma_min' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigma_min'] = self.model_wrap.sigmas[0].item()
|
||||
extra_params_kwargs['sigma_max'] = self.model_wrap.sigmas[-1].item()
|
||||
if 'n' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['n'] = steps
|
||||
else:
|
||||
extra_params_kwargs['sigmas'] = sigmas
|
||||
|
||||
self.last_latent = x
|
||||
samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args={
|
||||
'cond': conditioning,
|
||||
'image_cond': image_conditioning,
|
||||
'uncond': unconditional_conditioning,
|
||||
'cond_scale': p.cfg_scale
|
||||
}, disable=False, callback=self.callback_state, **extra_params_kwargs))
|
||||
|
||||
return samples
|
||||
|
||||
|
||||
78
modules/sd_samplers_common.py
Normal file
78
modules/sd_samplers_common.py
Normal file
@ -0,0 +1,78 @@
|
||||
from collections import namedtuple
|
||||
import numpy as np
|
||||
import torch
|
||||
from PIL import Image
|
||||
import torchsde._brownian.brownian_interval
|
||||
from modules import devices, processing, images, sd_vae_approx
|
||||
|
||||
from modules.shared import opts, state
|
||||
import modules.shared as shared
|
||||
|
||||
SamplerData = namedtuple('SamplerData', ['name', 'constructor', 'aliases', 'options'])
|
||||
|
||||
|
||||
def setup_img2img_steps(p, steps=None):
|
||||
if opts.img2img_fix_steps or steps is not None:
|
||||
requested_steps = (steps or p.steps)
|
||||
steps = int(requested_steps / min(p.denoising_strength, 0.999)) if p.denoising_strength > 0 else 0
|
||||
t_enc = requested_steps - 1
|
||||
else:
|
||||
steps = p.steps
|
||||
t_enc = int(min(p.denoising_strength, 0.999) * steps)
|
||||
|
||||
return steps, t_enc
|
||||
|
||||
|
||||
approximation_indexes = {"Full": 0, "Approx NN": 1, "Approx cheap": 2}
|
||||
|
||||
|
||||
def single_sample_to_image(sample, approximation=None):
|
||||
if approximation is None:
|
||||
approximation = approximation_indexes.get(opts.show_progress_type, 0)
|
||||
|
||||
if approximation == 2:
|
||||
x_sample = sd_vae_approx.cheap_approximation(sample)
|
||||
elif approximation == 1:
|
||||
x_sample = sd_vae_approx.model()(sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach()
|
||||
else:
|
||||
x_sample = processing.decode_first_stage(shared.sd_model, sample.unsqueeze(0))[0]
|
||||
|
||||
x_sample = torch.clamp((x_sample + 1.0) / 2.0, min=0.0, max=1.0)
|
||||
x_sample = 255. * np.moveaxis(x_sample.cpu().numpy(), 0, 2)
|
||||
x_sample = x_sample.astype(np.uint8)
|
||||
return Image.fromarray(x_sample)
|
||||
|
||||
|
||||
def sample_to_image(samples, index=0, approximation=None):
|
||||
return single_sample_to_image(samples[index], approximation)
|
||||
|
||||
|
||||
def samples_to_image_grid(samples, approximation=None):
|
||||
return images.image_grid([single_sample_to_image(sample, approximation) for sample in samples])
|
||||
|
||||
|
||||
def store_latent(decoded):
|
||||
state.current_latent = decoded
|
||||
|
||||
if opts.live_previews_enable and opts.show_progress_every_n_steps > 0 and shared.state.sampling_step % opts.show_progress_every_n_steps == 0:
|
||||
if not shared.parallel_processing_allowed:
|
||||
shared.state.assign_current_image(sample_to_image(decoded))
|
||||
|
||||
|
||||
class InterruptedException(BaseException):
|
||||
pass
|
||||
|
||||
|
||||
# MPS fix for randn in torchsde
|
||||
# XXX move this to separate file for MPS
|
||||
def torchsde_randn(size, dtype, device, seed):
|
||||
if device.type == 'mps':
|
||||
generator = torch.Generator(devices.cpu).manual_seed(int(seed))
|
||||
return torch.randn(size, dtype=dtype, device=devices.cpu, generator=generator).to(device)
|
||||
else:
|
||||
generator = torch.Generator(device).manual_seed(int(seed))
|
||||
return torch.randn(size, dtype=dtype, device=device, generator=generator)
|
||||
|
||||
|
||||
torchsde._brownian.brownian_interval._randn = torchsde_randn
|
||||
|
||||
160
modules/sd_samplers_compvis.py
Normal file
160
modules/sd_samplers_compvis.py
Normal file
@ -0,0 +1,160 @@
|
||||
import math
|
||||
import ldm.models.diffusion.ddim
|
||||
import ldm.models.diffusion.plms
|
||||
|
||||
import numpy as np
|
||||
import torch
|
||||
|
||||
from modules.shared import state
|
||||
from modules import sd_samplers_common, prompt_parser, shared
|
||||
|
||||
|
||||
samplers_data_compvis = [
|
||||
sd_samplers_common.SamplerData('DDIM', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.ddim.DDIMSampler, model), [], {}),
|
||||
sd_samplers_common.SamplerData('PLMS', lambda model: VanillaStableDiffusionSampler(ldm.models.diffusion.plms.PLMSSampler, model), [], {}),
|
||||
]
|
||||
|
||||
|
||||
class VanillaStableDiffusionSampler:
|
||||
def __init__(self, constructor, sd_model):
|
||||
self.sampler = constructor(sd_model)
|
||||
self.is_plms = hasattr(self.sampler, 'p_sample_plms')
|
||||
self.orig_p_sample_ddim = self.sampler.p_sample_plms if self.is_plms else self.sampler.p_sample_ddim
|
||||
self.mask = None
|
||||
self.nmask = None
|
||||
self.init_latent = None
|
||||
self.sampler_noises = None
|
||||
self.step = 0
|
||||
self.stop_at = None
|
||||
self.eta = None
|
||||
self.config = None
|
||||
self.last_latent = None
|
||||
|
||||
self.conditioning_key = sd_model.model.conditioning_key
|
||||
|
||||
def number_of_needed_noises(self, p):
|
||||
return 0
|
||||
|
||||
def launch_sampling(self, steps, func):
|
||||
state.sampling_steps = steps
|
||||
state.sampling_step = 0
|
||||
|
||||
try:
|
||||
return func()
|
||||
except sd_samplers_common.InterruptedException:
|
||||
return self.last_latent
|
||||
|
||||
def p_sample_ddim_hook(self, x_dec, cond, ts, unconditional_conditioning, *args, **kwargs):
|
||||
if state.interrupted or state.skipped:
|
||||
raise sd_samplers_common.InterruptedException
|
||||
|
||||
if self.stop_at is not None and self.step > self.stop_at:
|
||||
raise sd_samplers_common.InterruptedException
|
||||
|
||||
# Have to unwrap the inpainting conditioning here to perform pre-processing
|
||||
image_conditioning = None
|
||||
if isinstance(cond, dict):
|
||||
image_conditioning = cond["c_concat"][0]
|
||||
cond = cond["c_crossattn"][0]
|
||||
unconditional_conditioning = unconditional_conditioning["c_crossattn"][0]
|
||||
|
||||
conds_list, tensor = prompt_parser.reconstruct_multicond_batch(cond, self.step)
|
||||
unconditional_conditioning = prompt_parser.reconstruct_cond_batch(unconditional_conditioning, self.step)
|
||||
|
||||
assert all([len(conds) == 1 for conds in conds_list]), 'composition via AND is not supported for DDIM/PLMS samplers'
|
||||
cond = tensor
|
||||
|
||||
# for DDIM, shapes must match, we can't just process cond and uncond independently;
|
||||
# filling unconditional_conditioning with repeats of the last vector to match length is
|
||||
# not 100% correct but should work well enough
|
||||
if unconditional_conditioning.shape[1] < cond.shape[1]:
|
||||
last_vector = unconditional_conditioning[:, -1:]
|
||||
last_vector_repeated = last_vector.repeat([1, cond.shape[1] - unconditional_conditioning.shape[1], 1])
|
||||
unconditional_conditioning = torch.hstack([unconditional_conditioning, last_vector_repeated])
|
||||
elif unconditional_conditioning.shape[1] > cond.shape[1]:
|
||||
unconditional_conditioning = unconditional_conditioning[:, :cond.shape[1]]
|
||||
|
||||
if self.mask is not None:
|
||||
img_orig = self.sampler.model.q_sample(self.init_latent, ts)
|
||||
x_dec = img_orig * self.mask + self.nmask * x_dec
|
||||
|
||||
# Wrap the image conditioning back up since the DDIM code can accept the dict directly.
|
||||
# Note that they need to be lists because it just concatenates them later.
|
||||
if image_conditioning is not None:
|
||||
cond = {"c_concat": [image_conditioning], "c_crossattn": [cond]}
|
||||
unconditional_conditioning = {"c_concat": [image_conditioning], "c_crossattn": [unconditional_conditioning]}
|
||||
|
||||
res = self.orig_p_sample_ddim(x_dec, cond, ts, unconditional_conditioning=unconditional_conditioning, *args, **kwargs)
|
||||
|
||||
if self.mask is not None:
|
||||
self.last_latent = self.init_latent * self.mask + self.nmask * res[1]
|
||||
else:
|
||||
self.last_latent = res[1]
|
||||
|
||||
sd_samplers_common.store_latent(self.last_latent)
|
||||
|
||||
self.step += 1
|
||||
state.sampling_step = self.step
|
||||
shared.total_tqdm.update()
|
||||
|
||||
return res
|
||||
|
||||
def initialize(self, p):
|
||||
self.eta = p.eta if p.eta is not None else shared.opts.eta_ddim
|
||||
if self.eta != 0.0:
|
||||
p.extra_generation_params["Eta DDIM"] = self.eta
|
||||
|
||||
for fieldname in ['p_sample_ddim', 'p_sample_plms']:
|
||||
if hasattr(self.sampler, fieldname):
|
||||
setattr(self.sampler, fieldname, self.p_sample_ddim_hook)
|
||||
|
||||
self.mask = p.mask if hasattr(p, 'mask') else None
|
||||
self.nmask = p.nmask if hasattr(p, 'nmask') else None
|
||||
|
||||
def adjust_steps_if_invalid(self, p, num_steps):
|
||||
if (self.config.name == 'DDIM' and p.ddim_discretize == 'uniform') or (self.config.name == 'PLMS'):
|
||||
valid_step = 999 / (1000 // num_steps)
|
||||
if valid_step == math.floor(valid_step):
|
||||
return int(valid_step) + 1
|
||||
|
||||
return num_steps
|
||||
|
||||
def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None, image_conditioning=None):
|
||||
steps, t_enc = sd_samplers_common.setup_img2img_steps(p, steps)
|
||||
steps = self.adjust_steps_if_invalid(p, steps)
|
||||
self.initialize(p)
|
||||
|
||||
self.sampler.make_schedule(ddim_num_steps=steps, ddim_eta=self.eta, ddim_discretize=p.ddim_discretize, verbose=False)
|
||||
x1 = self.sampler.stochastic_encode(x, torch.tensor([t_enc] * int(x.shape[0])).to(shared.device), noise=noise)
|
||||
|
||||
self.init_latent = x
|
||||
self.last_latent = x
|
||||
self.step = 0
|
||||
|
||||
# Wrap the conditioning models with additional image conditioning for inpainting model
|
||||
if image_conditioning is not None:
|
||||
conditioning = {"c_concat": [image_conditioning], "c_crossattn": [conditioning]}
|
||||
unconditional_conditioning = {"c_concat": [image_conditioning], "c_crossattn": [unconditional_conditioning]}
|
||||
|
||||
samples = self.launch_sampling(t_enc + 1, lambda: self.sampler.decode(x1, conditioning, t_enc, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning))
|
||||
|
||||
return samples
|
||||
|
||||
def sample(self, p, x, conditioning, unconditional_conditioning, steps=None, image_conditioning=None):
|
||||
self.initialize(p)
|
||||
|
||||
self.init_latent = None
|
||||
self.last_latent = x
|
||||
self.step = 0
|
||||
|
||||
steps = self.adjust_steps_if_invalid(p, steps or p.steps)
|
||||
|
||||
# Wrap the conditioning models with additional image conditioning for inpainting model
|
||||
# dummy_for_plms is needed because PLMS code checks the first item in the dict to have the right shape
|
||||
if image_conditioning is not None:
|
||||
conditioning = {"dummy_for_plms": np.zeros((conditioning.shape[0],)), "c_crossattn": [conditioning], "c_concat": [image_conditioning]}
|
||||
unconditional_conditioning = {"c_crossattn": [unconditional_conditioning], "c_concat": [image_conditioning]}
|
||||
|
||||
samples_ddim = self.launch_sampling(steps, lambda: self.sampler.sample(S=steps, conditioning=conditioning, batch_size=int(x.shape[0]), shape=x[0].shape, verbose=False, unconditional_guidance_scale=p.cfg_scale, unconditional_conditioning=unconditional_conditioning, x_T=x, eta=self.eta)[0])
|
||||
|
||||
return samples_ddim
|
||||
298
modules/sd_samplers_kdiffusion.py
Normal file
298
modules/sd_samplers_kdiffusion.py
Normal file
@ -0,0 +1,298 @@
|
||||
from collections import deque
|
||||
import torch
|
||||
import inspect
|
||||
import k_diffusion.sampling
|
||||
from modules import prompt_parser, devices, sd_samplers_common
|
||||
|
||||
from modules.shared import opts, state
|
||||
import modules.shared as shared
|
||||
from modules.script_callbacks import CFGDenoiserParams, cfg_denoiser_callback
|
||||
|
||||
samplers_k_diffusion = [
|
||||
('Euler a', 'sample_euler_ancestral', ['k_euler_a', 'k_euler_ancestral'], {}),
|
||||
('Euler', 'sample_euler', ['k_euler'], {}),
|
||||
('LMS', 'sample_lms', ['k_lms'], {}),
|
||||
('Heun', 'sample_heun', ['k_heun'], {}),
|
||||
('DPM2', 'sample_dpm_2', ['k_dpm_2'], {'discard_next_to_last_sigma': True}),
|
||||
('DPM2 a', 'sample_dpm_2_ancestral', ['k_dpm_2_a'], {'discard_next_to_last_sigma': True}),
|
||||
('DPM++ 2S a', 'sample_dpmpp_2s_ancestral', ['k_dpmpp_2s_a'], {}),
|
||||
('DPM++ 2M', 'sample_dpmpp_2m', ['k_dpmpp_2m'], {}),
|
||||
('DPM++ SDE', 'sample_dpmpp_sde', ['k_dpmpp_sde'], {}),
|
||||
('DPM fast', 'sample_dpm_fast', ['k_dpm_fast'], {}),
|
||||
('DPM adaptive', 'sample_dpm_adaptive', ['k_dpm_ad'], {}),
|
||||
('LMS Karras', 'sample_lms', ['k_lms_ka'], {'scheduler': 'karras'}),
|
||||
('DPM2 Karras', 'sample_dpm_2', ['k_dpm_2_ka'], {'scheduler': 'karras', 'discard_next_to_last_sigma': True}),
|
||||
('DPM2 a Karras', 'sample_dpm_2_ancestral', ['k_dpm_2_a_ka'], {'scheduler': 'karras', 'discard_next_to_last_sigma': True}),
|
||||
('DPM++ 2S a Karras', 'sample_dpmpp_2s_ancestral', ['k_dpmpp_2s_a_ka'], {'scheduler': 'karras'}),
|
||||
('DPM++ 2M Karras', 'sample_dpmpp_2m', ['k_dpmpp_2m_ka'], {'scheduler': 'karras'}),
|
||||
('DPM++ SDE Karras', 'sample_dpmpp_sde', ['k_dpmpp_sde_ka'], {'scheduler': 'karras'}),
|
||||
]
|
||||
|
||||
samplers_data_k_diffusion = [
|
||||
sd_samplers_common.SamplerData(label, lambda model, funcname=funcname: KDiffusionSampler(funcname, model), aliases, options)
|
||||
for label, funcname, aliases, options in samplers_k_diffusion
|
||||
if hasattr(k_diffusion.sampling, funcname)
|
||||
]
|
||||
|
||||
sampler_extra_params = {
|
||||
'sample_euler': ['s_churn', 's_tmin', 's_tmax', 's_noise'],
|
||||
'sample_heun': ['s_churn', 's_tmin', 's_tmax', 's_noise'],
|
||||
'sample_dpm_2': ['s_churn', 's_tmin', 's_tmax', 's_noise'],
|
||||
}
|
||||
|
||||
|
||||
class CFGDenoiser(torch.nn.Module):
|
||||
"""
|
||||
Classifier free guidance denoiser. A wrapper for stable diffusion model (specifically for unet)
|
||||
that can take a noisy picture and produce a noise-free picture using two guidances (prompts)
|
||||
instead of one. Originally, the second prompt is just an empty string, but we use non-empty
|
||||
negative prompt.
|
||||
"""
|
||||
|
||||
def __init__(self, model):
|
||||
super().__init__()
|
||||
self.inner_model = model
|
||||
self.mask = None
|
||||
self.nmask = None
|
||||
self.init_latent = None
|
||||
self.step = 0
|
||||
|
||||
def combine_denoised(self, x_out, conds_list, uncond, cond_scale):
|
||||
denoised_uncond = x_out[-uncond.shape[0]:]
|
||||
denoised = torch.clone(denoised_uncond)
|
||||
|
||||
for i, conds in enumerate(conds_list):
|
||||
for cond_index, weight in conds:
|
||||
denoised[i] += (x_out[cond_index] - denoised_uncond[i]) * (weight * cond_scale)
|
||||
|
||||
return denoised
|
||||
|
||||
def forward(self, x, sigma, uncond, cond, cond_scale, image_cond):
|
||||
if state.interrupted or state.skipped:
|
||||
raise sd_samplers_common.InterruptedException
|
||||
|
||||
conds_list, tensor = prompt_parser.reconstruct_multicond_batch(cond, self.step)
|
||||
uncond = prompt_parser.reconstruct_cond_batch(uncond, self.step)
|
||||
|
||||
batch_size = len(conds_list)
|
||||
repeats = [len(conds_list[i]) for i in range(batch_size)]
|
||||
|
||||
x_in = torch.cat([torch.stack([x[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [x])
|
||||
image_cond_in = torch.cat([torch.stack([image_cond[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [image_cond])
|
||||
sigma_in = torch.cat([torch.stack([sigma[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [sigma])
|
||||
|
||||
denoiser_params = CFGDenoiserParams(x_in, image_cond_in, sigma_in, state.sampling_step, state.sampling_steps)
|
||||
cfg_denoiser_callback(denoiser_params)
|
||||
x_in = denoiser_params.x
|
||||
image_cond_in = denoiser_params.image_cond
|
||||
sigma_in = denoiser_params.sigma
|
||||
|
||||
if tensor.shape[1] == uncond.shape[1]:
|
||||
cond_in = torch.cat([tensor, uncond])
|
||||
|
||||
if shared.batch_cond_uncond:
|
||||
x_out = self.inner_model(x_in, sigma_in, cond={"c_crossattn": [cond_in], "c_concat": [image_cond_in]})
|
||||
else:
|
||||
x_out = torch.zeros_like(x_in)
|
||||
for batch_offset in range(0, x_out.shape[0], batch_size):
|
||||
a = batch_offset
|
||||
b = a + batch_size
|
||||
x_out[a:b] = self.inner_model(x_in[a:b], sigma_in[a:b], cond={"c_crossattn": [cond_in[a:b]], "c_concat": [image_cond_in[a:b]]})
|
||||
else:
|
||||
x_out = torch.zeros_like(x_in)
|
||||
batch_size = batch_size*2 if shared.batch_cond_uncond else batch_size
|
||||
for batch_offset in range(0, tensor.shape[0], batch_size):
|
||||
a = batch_offset
|
||||
b = min(a + batch_size, tensor.shape[0])
|
||||
x_out[a:b] = self.inner_model(x_in[a:b], sigma_in[a:b], cond={"c_crossattn": [tensor[a:b]], "c_concat": [image_cond_in[a:b]]})
|
||||
|
||||
x_out[-uncond.shape[0]:] = self.inner_model(x_in[-uncond.shape[0]:], sigma_in[-uncond.shape[0]:], cond={"c_crossattn": [uncond], "c_concat": [image_cond_in[-uncond.shape[0]:]]})
|
||||
|
||||
devices.test_for_nans(x_out, "unet")
|
||||
|
||||
if opts.live_preview_content == "Prompt":
|
||||
sd_samplers_common.store_latent(x_out[0:uncond.shape[0]])
|
||||
elif opts.live_preview_content == "Negative prompt":
|
||||
sd_samplers_common.store_latent(x_out[-uncond.shape[0]:])
|
||||
|
||||
denoised = self.combine_denoised(x_out, conds_list, uncond, cond_scale)
|
||||
|
||||
if self.mask is not None:
|
||||
denoised = self.init_latent * self.mask + self.nmask * denoised
|
||||
|
||||
self.step += 1
|
||||
|
||||
return denoised
|
||||
|
||||
|
||||
class TorchHijack:
|
||||
def __init__(self, sampler_noises):
|
||||
# Using a deque to efficiently receive the sampler_noises in the same order as the previous index-based
|
||||
# implementation.
|
||||
self.sampler_noises = deque(sampler_noises)
|
||||
|
||||
def __getattr__(self, item):
|
||||
if item == 'randn_like':
|
||||
return self.randn_like
|
||||
|
||||
if hasattr(torch, item):
|
||||
return getattr(torch, item)
|
||||
|
||||
raise AttributeError("'{}' object has no attribute '{}'".format(type(self).__name__, item))
|
||||
|
||||
def randn_like(self, x):
|
||||
if self.sampler_noises:
|
||||
noise = self.sampler_noises.popleft()
|
||||
if noise.shape == x.shape:
|
||||
return noise
|
||||
|
||||
if x.device.type == 'mps':
|
||||
return torch.randn_like(x, device=devices.cpu).to(x.device)
|
||||
else:
|
||||
return torch.randn_like(x)
|
||||
|
||||
|
||||
class KDiffusionSampler:
|
||||
def __init__(self, funcname, sd_model):
|
||||
denoiser = k_diffusion.external.CompVisVDenoiser if sd_model.parameterization == "v" else k_diffusion.external.CompVisDenoiser
|
||||
|
||||
self.model_wrap = denoiser(sd_model, quantize=shared.opts.enable_quantization)
|
||||
self.funcname = funcname
|
||||
self.func = getattr(k_diffusion.sampling, self.funcname)
|
||||
self.extra_params = sampler_extra_params.get(funcname, [])
|
||||
self.model_wrap_cfg = CFGDenoiser(self.model_wrap)
|
||||
self.sampler_noises = None
|
||||
self.stop_at = None
|
||||
self.eta = None
|
||||
self.config = None
|
||||
self.last_latent = None
|
||||
|
||||
self.conditioning_key = sd_model.model.conditioning_key
|
||||
|
||||
def callback_state(self, d):
|
||||
step = d['i']
|
||||
latent = d["denoised"]
|
||||
if opts.live_preview_content == "Combined":
|
||||
sd_samplers_common.store_latent(latent)
|
||||
self.last_latent = latent
|
||||
|
||||
if self.stop_at is not None and step > self.stop_at:
|
||||
raise sd_samplers_common.InterruptedException
|
||||
|
||||
state.sampling_step = step
|
||||
shared.total_tqdm.update()
|
||||
|
||||
def launch_sampling(self, steps, func):
|
||||
state.sampling_steps = steps
|
||||
state.sampling_step = 0
|
||||
|
||||
try:
|
||||
return func()
|
||||
except sd_samplers_common.InterruptedException:
|
||||
return self.last_latent
|
||||
|
||||
def number_of_needed_noises(self, p):
|
||||
return p.steps
|
||||
|
||||
def initialize(self, p):
|
||||
self.model_wrap_cfg.mask = p.mask if hasattr(p, 'mask') else None
|
||||
self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None
|
||||
self.model_wrap_cfg.step = 0
|
||||
self.eta = p.eta if p.eta is not None else opts.eta_ancestral
|
||||
|
||||
k_diffusion.sampling.torch = TorchHijack(self.sampler_noises if self.sampler_noises is not None else [])
|
||||
|
||||
extra_params_kwargs = {}
|
||||
for param_name in self.extra_params:
|
||||
if hasattr(p, param_name) and param_name in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs[param_name] = getattr(p, param_name)
|
||||
|
||||
if 'eta' in inspect.signature(self.func).parameters:
|
||||
if self.eta != 1.0:
|
||||
p.extra_generation_params["Eta"] = self.eta
|
||||
|
||||
extra_params_kwargs['eta'] = self.eta
|
||||
|
||||
return extra_params_kwargs
|
||||
|
||||
def get_sigmas(self, p, steps):
|
||||
discard_next_to_last_sigma = self.config is not None and self.config.options.get('discard_next_to_last_sigma', False)
|
||||
if opts.always_discard_next_to_last_sigma and not discard_next_to_last_sigma:
|
||||
discard_next_to_last_sigma = True
|
||||
p.extra_generation_params["Discard penultimate sigma"] = True
|
||||
|
||||
steps += 1 if discard_next_to_last_sigma else 0
|
||||
|
||||
if p.sampler_noise_scheduler_override:
|
||||
sigmas = p.sampler_noise_scheduler_override(steps)
|
||||
elif self.config is not None and self.config.options.get('scheduler', None) == 'karras':
|
||||
sigma_min, sigma_max = (0.1, 10) if opts.use_old_karras_scheduler_sigmas else (self.model_wrap.sigmas[0].item(), self.model_wrap.sigmas[-1].item())
|
||||
|
||||
sigmas = k_diffusion.sampling.get_sigmas_karras(n=steps, sigma_min=sigma_min, sigma_max=sigma_max, device=shared.device)
|
||||
else:
|
||||
sigmas = self.model_wrap.get_sigmas(steps)
|
||||
|
||||
if discard_next_to_last_sigma:
|
||||
sigmas = torch.cat([sigmas[:-2], sigmas[-1:]])
|
||||
|
||||
return sigmas
|
||||
|
||||
def sample_img2img(self, p, x, noise, conditioning, unconditional_conditioning, steps=None, image_conditioning=None):
|
||||
steps, t_enc = sd_samplers_common.setup_img2img_steps(p, steps)
|
||||
|
||||
sigmas = self.get_sigmas(p, steps)
|
||||
|
||||
sigma_sched = sigmas[steps - t_enc - 1:]
|
||||
xi = x + noise * sigma_sched[0]
|
||||
|
||||
extra_params_kwargs = self.initialize(p)
|
||||
if 'sigma_min' in inspect.signature(self.func).parameters:
|
||||
## last sigma is zero which isn't allowed by DPM Fast & Adaptive so taking value before last
|
||||
extra_params_kwargs['sigma_min'] = sigma_sched[-2]
|
||||
if 'sigma_max' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigma_max'] = sigma_sched[0]
|
||||
if 'n' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['n'] = len(sigma_sched) - 1
|
||||
if 'sigma_sched' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigma_sched'] = sigma_sched
|
||||
if 'sigmas' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigmas'] = sigma_sched
|
||||
|
||||
self.model_wrap_cfg.init_latent = x
|
||||
self.last_latent = x
|
||||
|
||||
samples = self.launch_sampling(t_enc + 1, lambda: self.func(self.model_wrap_cfg, xi, extra_args={
|
||||
'cond': conditioning,
|
||||
'image_cond': image_conditioning,
|
||||
'uncond': unconditional_conditioning,
|
||||
'cond_scale': p.cfg_scale
|
||||
}, disable=False, callback=self.callback_state, **extra_params_kwargs))
|
||||
|
||||
return samples
|
||||
|
||||
def sample(self, p, x, conditioning, unconditional_conditioning, steps=None, image_conditioning = None):
|
||||
steps = steps or p.steps
|
||||
|
||||
sigmas = self.get_sigmas(p, steps)
|
||||
|
||||
x = x * sigmas[0]
|
||||
|
||||
extra_params_kwargs = self.initialize(p)
|
||||
if 'sigma_min' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['sigma_min'] = self.model_wrap.sigmas[0].item()
|
||||
extra_params_kwargs['sigma_max'] = self.model_wrap.sigmas[-1].item()
|
||||
if 'n' in inspect.signature(self.func).parameters:
|
||||
extra_params_kwargs['n'] = steps
|
||||
else:
|
||||
extra_params_kwargs['sigmas'] = sigmas
|
||||
|
||||
self.last_latent = x
|
||||
samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args={
|
||||
'cond': conditioning,
|
||||
'image_cond': image_conditioning,
|
||||
'uncond': unconditional_conditioning,
|
||||
'cond_scale': p.cfg_scale
|
||||
}, disable=False, callback=self.callback_state, **extra_params_kwargs))
|
||||
|
||||
return samples
|
||||
|
||||
@ -127,12 +127,13 @@ restricted_opts = {
|
||||
ui_reorder_categories = [
|
||||
"inpaint",
|
||||
"sampler",
|
||||
"checkboxes",
|
||||
"hires_fix",
|
||||
"dimensions",
|
||||
"cfg",
|
||||
"seed",
|
||||
"checkboxes",
|
||||
"hires_fix",
|
||||
"batch",
|
||||
"override_settings",
|
||||
"scripts",
|
||||
]
|
||||
|
||||
@ -346,10 +347,10 @@ options_templates.update(options_section(('saving-paths', "Paths for saving"), {
|
||||
}))
|
||||
|
||||
options_templates.update(options_section(('saving-to-dirs', "Saving to a directory"), {
|
||||
"save_to_dirs": OptionInfo(False, "Save images to a subdirectory"),
|
||||
"grid_save_to_dirs": OptionInfo(False, "Save grids to a subdirectory"),
|
||||
"save_to_dirs": OptionInfo(True, "Save images to a subdirectory"),
|
||||
"grid_save_to_dirs": OptionInfo(True, "Save grids to a subdirectory"),
|
||||
"use_save_to_dirs_for_ui": OptionInfo(False, "When using \"Save\" button, save images to a subdirectory"),
|
||||
"directories_filename_pattern": OptionInfo("", "Directory name pattern", component_args=hide_dirs),
|
||||
"directories_filename_pattern": OptionInfo("[date]", "Directory name pattern", component_args=hide_dirs),
|
||||
"directories_max_prompt_words": OptionInfo(8, "Max prompt words for [prompt_words] pattern", gr.Slider, {"minimum": 1, "maximum": 20, "step": 1, **hide_dirs}),
|
||||
}))
|
||||
|
||||
@ -405,7 +406,6 @@ options_templates.update(options_section(('sd', "Stable Diffusion"), {
|
||||
"enable_batch_seeds": OptionInfo(True, "Make K-diffusion samplers produce same images in a batch as when making a single image"),
|
||||
"comma_padding_backtrack": OptionInfo(20, "Increase coherency by padding from the last comma within n tokens when using more than 75 tokens", gr.Slider, {"minimum": 0, "maximum": 74, "step": 1 }),
|
||||
"CLIP_stop_at_last_layers": OptionInfo(1, "Clip skip", gr.Slider, {"minimum": 1, "maximum": 12, "step": 1}),
|
||||
"extra_networks_default_multiplier": OptionInfo(1.0, "Multiplier for extra networks", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}),
|
||||
"upcast_attn": OptionInfo(False, "Upcast cross attention layer to float32"),
|
||||
}))
|
||||
|
||||
@ -431,7 +431,9 @@ options_templates.update(options_section(('interrogate', "Interrogate Options"),
|
||||
}))
|
||||
|
||||
options_templates.update(options_section(('extra_networks', "Extra Networks"), {
|
||||
"extra_networks_default_view": OptionInfo("cards", "Default view for Extra Networks", gr.Dropdown, { "choices": ["cards", "thumbs"] }),
|
||||
"extra_networks_default_view": OptionInfo("cards", "Default view for Extra Networks", gr.Dropdown, {"choices": ["cards", "thumbs"]}),
|
||||
"extra_networks_default_multiplier": OptionInfo(1.0, "Multiplier for extra networks", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}),
|
||||
"sd_hypernetwork": OptionInfo("None", "Add hypernetwork to prompt", gr.Dropdown, lambda: {"choices": [""] + [x for x in hypernetworks.keys()]}, refresh=reload_hypernetworks),
|
||||
}))
|
||||
|
||||
options_templates.update(options_section(('ui', "User interface"), {
|
||||
@ -439,7 +441,7 @@ options_templates.update(options_section(('ui', "User interface"), {
|
||||
"do_not_show_images": OptionInfo(False, "Do not show any images in results for web"),
|
||||
"add_model_hash_to_info": OptionInfo(True, "Add model hash to generation information"),
|
||||
"add_model_name_to_info": OptionInfo(True, "Add model name to generation information"),
|
||||
"disable_weights_auto_swap": OptionInfo(False, "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint."),
|
||||
"disable_weights_auto_swap": OptionInfo(True, "When reading generation parameters from text into UI (from PNG info or pasted text), do not change the selected model/checkpoint."),
|
||||
"send_seed": OptionInfo(True, "Send seed when sending prompt or image to other interface"),
|
||||
"send_size": OptionInfo(True, "Send size when sending prompt or image to another interface"),
|
||||
"font": OptionInfo("", "Font for image grids that have text"),
|
||||
@ -604,11 +606,37 @@ class Options:
|
||||
|
||||
self.data_labels = {k: v for k, v in sorted(settings_items, key=lambda x: section_ids[x[1].section])}
|
||||
|
||||
def cast_value(self, key, value):
|
||||
"""casts an arbitrary to the same type as this setting's value with key
|
||||
Example: cast_value("eta_noise_seed_delta", "12") -> returns 12 (an int rather than str)
|
||||
"""
|
||||
|
||||
if value is None:
|
||||
return None
|
||||
|
||||
default_value = self.data_labels[key].default
|
||||
if default_value is None:
|
||||
default_value = getattr(self, key, None)
|
||||
if default_value is None:
|
||||
return None
|
||||
|
||||
expected_type = type(default_value)
|
||||
if expected_type == bool and value == "False":
|
||||
value = False
|
||||
else:
|
||||
value = expected_type(value)
|
||||
|
||||
return value
|
||||
|
||||
|
||||
|
||||
opts = Options()
|
||||
if os.path.exists(config_filename):
|
||||
opts.load(config_filename)
|
||||
|
||||
settings_components = None
|
||||
"""assinged from ui.py, a mapping on setting anmes to gradio components repsponsible for those settings"""
|
||||
|
||||
latent_upscale_default_mode = "Latent"
|
||||
latent_upscale_modes = {
|
||||
"Latent": {"mode": "bilinear", "antialias": False},
|
||||
|
||||
@ -112,6 +112,7 @@ class EmbeddingDatabase:
|
||||
self.skipped_embeddings = {}
|
||||
self.expected_shape = -1
|
||||
self.embedding_dirs = {}
|
||||
self.previously_displayed_embeddings = ()
|
||||
|
||||
def add_embedding_dir(self, path):
|
||||
self.embedding_dirs[path] = DirWithTextualInversionEmbeddings(path)
|
||||
@ -228,9 +229,12 @@ class EmbeddingDatabase:
|
||||
self.load_from_dir(embdir)
|
||||
embdir.update()
|
||||
|
||||
print(f"Textual inversion embeddings loaded({len(self.word_embeddings)}): {', '.join(self.word_embeddings.keys())}")
|
||||
if len(self.skipped_embeddings) > 0:
|
||||
print(f"Textual inversion embeddings skipped({len(self.skipped_embeddings)}): {', '.join(self.skipped_embeddings.keys())}")
|
||||
displayed_embeddings = (tuple(self.word_embeddings.keys()), tuple(self.skipped_embeddings.keys()))
|
||||
if self.previously_displayed_embeddings != displayed_embeddings:
|
||||
self.previously_displayed_embeddings = displayed_embeddings
|
||||
print(f"Textual inversion embeddings loaded({len(self.word_embeddings)}): {', '.join(self.word_embeddings.keys())}")
|
||||
if len(self.skipped_embeddings) > 0:
|
||||
print(f"Textual inversion embeddings skipped({len(self.skipped_embeddings)}): {', '.join(self.skipped_embeddings.keys())}")
|
||||
|
||||
def find_embedding_at_position(self, tokens, offset):
|
||||
token = tokens[offset]
|
||||
|
||||
@ -1,5 +1,6 @@
|
||||
import modules.scripts
|
||||
from modules import sd_samplers
|
||||
from modules.generation_parameters_copypaste import create_override_settings_dict
|
||||
from modules.processing import StableDiffusionProcessing, Processed, StableDiffusionProcessingTxt2Img, \
|
||||
StableDiffusionProcessingImg2Img, process_images
|
||||
from modules.shared import opts, cmd_opts
|
||||
@ -8,7 +9,9 @@ import modules.processing as processing
|
||||
from modules.ui import plaintext_to_html
|
||||
|
||||
|
||||
def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_index: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, 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, *args):
|
||||
def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_index: int, restore_faces: bool, tiling: bool, n_iter: int, batch_size: int, cfg_scale: float, seed: int, subseed: int, subseed_strength: float, seed_resize_from_h: int, seed_resize_from_w: int, seed_enable_extras: bool, 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, override_settings_texts, *args):
|
||||
override_settings = create_override_settings_dict(override_settings_texts)
|
||||
|
||||
p = StableDiffusionProcessingTxt2Img(
|
||||
sd_model=shared.sd_model,
|
||||
outpath_samples=opts.outdir_samples or opts.outdir_txt2img_samples,
|
||||
@ -38,6 +41,7 @@ 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,
|
||||
override_settings=override_settings,
|
||||
)
|
||||
|
||||
p.scripts = modules.scripts.scripts_txt2img
|
||||
|
||||
@ -380,6 +380,7 @@ def apply_setting(key, value):
|
||||
opts.save(shared.config_filename)
|
||||
return getattr(opts, key)
|
||||
|
||||
|
||||
def create_refresh_button(refresh_component, refresh_method, refreshed_args, elem_id):
|
||||
def refresh():
|
||||
refresh_method()
|
||||
@ -433,6 +434,18 @@ def get_value_for_setting(key):
|
||||
return gr.update(value=value, **args)
|
||||
|
||||
|
||||
def create_override_settings_dropdown(tabname, row):
|
||||
dropdown = gr.Dropdown([], label="Override settings", visible=False, elem_id=f"{tabname}_override_settings", multiselect=True)
|
||||
|
||||
dropdown.change(
|
||||
fn=lambda x: gr.Dropdown.update(visible=len(x) > 0),
|
||||
inputs=[dropdown],
|
||||
outputs=[dropdown],
|
||||
)
|
||||
|
||||
return dropdown
|
||||
|
||||
|
||||
def create_ui():
|
||||
import modules.img2img
|
||||
import modules.txt2img
|
||||
@ -503,6 +516,10 @@ def create_ui():
|
||||
batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="txt2img_batch_count")
|
||||
batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="txt2img_batch_size")
|
||||
|
||||
elif category == "override_settings":
|
||||
with FormRow(elem_id="txt2img_override_settings_row") as row:
|
||||
override_settings = create_override_settings_dropdown('txt2img', row)
|
||||
|
||||
elif category == "scripts":
|
||||
with FormGroup(elem_id="txt2img_script_container"):
|
||||
custom_inputs = modules.scripts.scripts_txt2img.setup_ui()
|
||||
@ -524,7 +541,6 @@ def create_ui():
|
||||
)
|
||||
|
||||
txt2img_gallery, generation_info, html_info, html_log = create_output_panel("txt2img", opts.outdir_txt2img_samples)
|
||||
parameters_copypaste.bind_buttons({"txt2img": txt2img_paste}, None, txt2img_prompt)
|
||||
|
||||
connect_reuse_seed(seed, reuse_seed, generation_info, dummy_component, is_subseed=False)
|
||||
connect_reuse_seed(subseed, reuse_subseed, generation_info, dummy_component, is_subseed=True)
|
||||
@ -555,6 +571,7 @@ def create_ui():
|
||||
hr_second_pass_steps,
|
||||
hr_resize_x,
|
||||
hr_resize_y,
|
||||
override_settings,
|
||||
] + custom_inputs,
|
||||
|
||||
outputs=[
|
||||
@ -615,6 +632,9 @@ def create_ui():
|
||||
*modules.scripts.scripts_txt2img.infotext_fields
|
||||
]
|
||||
parameters_copypaste.add_paste_fields("txt2img", None, txt2img_paste_fields)
|
||||
parameters_copypaste.register_paste_params_button(parameters_copypaste.ParamBinding(
|
||||
paste_button=txt2img_paste, tabname="txt2img", source_text_component=txt2img_prompt, source_image_component=None, override_settings_component=override_settings,
|
||||
))
|
||||
|
||||
txt2img_preview_params = [
|
||||
txt2img_prompt,
|
||||
@ -762,6 +782,10 @@ def create_ui():
|
||||
batch_count = gr.Slider(minimum=1, step=1, label='Batch count', value=1, elem_id="img2img_batch_count")
|
||||
batch_size = gr.Slider(minimum=1, maximum=8, step=1, label='Batch size', value=1, elem_id="img2img_batch_size")
|
||||
|
||||
elif category == "override_settings":
|
||||
with FormRow(elem_id="img2img_override_settings_row") as row:
|
||||
override_settings = create_override_settings_dropdown('img2img', row)
|
||||
|
||||
elif category == "scripts":
|
||||
with FormGroup(elem_id="img2img_script_container"):
|
||||
custom_inputs = modules.scripts.scripts_img2img.setup_ui()
|
||||
@ -796,7 +820,6 @@ def create_ui():
|
||||
)
|
||||
|
||||
img2img_gallery, generation_info, html_info, html_log = create_output_panel("img2img", opts.outdir_img2img_samples)
|
||||
parameters_copypaste.bind_buttons({"img2img": img2img_paste}, None, img2img_prompt)
|
||||
|
||||
connect_reuse_seed(seed, reuse_seed, generation_info, dummy_component, is_subseed=False)
|
||||
connect_reuse_seed(subseed, reuse_subseed, generation_info, dummy_component, is_subseed=True)
|
||||
@ -849,7 +872,8 @@ def create_ui():
|
||||
inpainting_mask_invert,
|
||||
img2img_batch_input_dir,
|
||||
img2img_batch_output_dir,
|
||||
img2img_batch_inpaint_mask_dir
|
||||
img2img_batch_inpaint_mask_dir,
|
||||
override_settings,
|
||||
] + custom_inputs,
|
||||
outputs=[
|
||||
img2img_gallery,
|
||||
@ -937,6 +961,9 @@ def create_ui():
|
||||
]
|
||||
parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields)
|
||||
parameters_copypaste.add_paste_fields("inpaint", init_img_with_mask, img2img_paste_fields)
|
||||
parameters_copypaste.register_paste_params_button(parameters_copypaste.ParamBinding(
|
||||
paste_button=img2img_paste, tabname="img2img", source_text_component=img2img_prompt, source_image_component=None, override_settings_component=override_settings,
|
||||
))
|
||||
|
||||
modules.scripts.scripts_current = None
|
||||
|
||||
@ -954,7 +981,11 @@ def create_ui():
|
||||
html2 = gr.HTML()
|
||||
with gr.Row():
|
||||
buttons = parameters_copypaste.create_buttons(["txt2img", "img2img", "inpaint", "extras"])
|
||||
parameters_copypaste.bind_buttons(buttons, image, generation_info)
|
||||
|
||||
for tabname, button in buttons.items():
|
||||
parameters_copypaste.register_paste_params_button(parameters_copypaste.ParamBinding(
|
||||
paste_button=button, tabname=tabname, source_text_component=generation_info, source_image_component=image,
|
||||
))
|
||||
|
||||
image.change(
|
||||
fn=wrap_gradio_call(modules.extras.run_pnginfo),
|
||||
@ -1363,6 +1394,7 @@ def create_ui():
|
||||
|
||||
components = []
|
||||
component_dict = {}
|
||||
shared.settings_components = component_dict
|
||||
|
||||
script_callbacks.ui_settings_callback()
|
||||
opts.reorder()
|
||||
@ -1529,8 +1561,7 @@ def create_ui():
|
||||
component = create_setting_component(k, is_quicksettings=True)
|
||||
component_dict[k] = component
|
||||
|
||||
parameters_copypaste.integrate_settings_paste_fields(component_dict)
|
||||
parameters_copypaste.run_bind()
|
||||
parameters_copypaste.connect_paste_params_buttons()
|
||||
|
||||
with gr.Tabs(elem_id="tabs") as tabs:
|
||||
for interface, label, ifid in interfaces:
|
||||
@ -1560,6 +1591,14 @@ def create_ui():
|
||||
outputs=[component, text_settings],
|
||||
)
|
||||
|
||||
button_set_checkpoint = gr.Button('Change checkpoint', elem_id='change_checkpoint', visible=False)
|
||||
button_set_checkpoint.click(
|
||||
fn=lambda value, _: run_settings_single(value, key='sd_model_checkpoint'),
|
||||
_js="function(v){ var res = desiredCheckpointName; desiredCheckpointName = ''; return [res || v, null]; }",
|
||||
inputs=[component_dict['sd_model_checkpoint'], dummy_component],
|
||||
outputs=[component_dict['sd_model_checkpoint'], text_settings],
|
||||
)
|
||||
|
||||
component_keys = [k for k in opts.data_labels.keys() if k in component_dict]
|
||||
|
||||
def get_settings_values():
|
||||
@ -1692,14 +1731,14 @@ def create_ui():
|
||||
|
||||
|
||||
def reload_javascript():
|
||||
head = f'<script type="text/javascript" src="file={os.path.abspath("script.js")}"></script>\n'
|
||||
head = f'<script type="text/javascript" src="file={os.path.abspath("script.js")}?{os.path.getmtime("script.js")}"></script>\n'
|
||||
|
||||
inline = f"{localization.localization_js(shared.opts.localization)};"
|
||||
if cmd_opts.theme is not None:
|
||||
inline += f"set_theme('{cmd_opts.theme}');"
|
||||
|
||||
for script in modules.scripts.list_scripts("javascript", ".js"):
|
||||
head += f'<script type="text/javascript" src="file={script.path}"></script>\n'
|
||||
head += f'<script type="text/javascript" src="file={script.path}?{os.path.getmtime(script.path)}"></script>\n'
|
||||
|
||||
head += f'<script type="text/javascript">{inline}</script>\n'
|
||||
|
||||
|
||||
@ -198,5 +198,9 @@ Requested path was: {f}
|
||||
html_info = gr.HTML(elem_id=f'html_info_{tabname}')
|
||||
html_log = gr.HTML(elem_id=f'html_log_{tabname}')
|
||||
|
||||
parameters_copypaste.bind_buttons(buttons, result_gallery, "txt2img" if tabname == "txt2img" else None)
|
||||
for paste_tabname, paste_button in buttons.items():
|
||||
parameters_copypaste.register_paste_params_button(parameters_copypaste.ParamBinding(
|
||||
paste_button=paste_button, tabname=paste_tabname, source_tabname="txt2img" if tabname == "txt2img" else None, source_image_component=result_gallery
|
||||
))
|
||||
|
||||
return result_gallery, generation_info if tabname != "extras" else html_info_x, html_info, html_log
|
||||
|
||||
@ -13,7 +13,7 @@ import shutil
|
||||
import errno
|
||||
|
||||
from modules import extensions, shared, paths
|
||||
|
||||
from modules.call_queue import wrap_gradio_gpu_call
|
||||
|
||||
available_extensions = {"extensions": []}
|
||||
|
||||
@ -50,12 +50,17 @@ def apply_and_restart(disable_list, update_list):
|
||||
shared.state.need_restart = True
|
||||
|
||||
|
||||
def check_updates():
|
||||
def check_updates(id_task, disable_list):
|
||||
check_access()
|
||||
|
||||
for ext in extensions.extensions:
|
||||
if ext.remote is None:
|
||||
continue
|
||||
disabled = json.loads(disable_list)
|
||||
assert type(disabled) == list, f"wrong disable_list data for apply_and_restart: {disable_list}"
|
||||
|
||||
exts = [ext for ext in extensions.extensions if ext.remote is not None and ext.name not in disabled]
|
||||
shared.state.job_count = len(exts)
|
||||
|
||||
for ext in exts:
|
||||
shared.state.textinfo = ext.name
|
||||
|
||||
try:
|
||||
ext.check_updates()
|
||||
@ -63,7 +68,9 @@ def check_updates():
|
||||
print(f"Error checking updates for {ext.name}:", file=sys.stderr)
|
||||
print(traceback.format_exc(), file=sys.stderr)
|
||||
|
||||
return extension_table()
|
||||
shared.state.nextjob()
|
||||
|
||||
return extension_table(), ""
|
||||
|
||||
|
||||
def extension_table():
|
||||
@ -273,12 +280,13 @@ def create_ui():
|
||||
with gr.Tabs(elem_id="tabs_extensions") as tabs:
|
||||
with gr.TabItem("Installed"):
|
||||
|
||||
with gr.Row():
|
||||
with gr.Row(elem_id="extensions_installed_top"):
|
||||
apply = gr.Button(value="Apply and restart UI", variant="primary")
|
||||
check = gr.Button(value="Check for updates")
|
||||
extensions_disabled_list = gr.Text(elem_id="extensions_disabled_list", visible=False).style(container=False)
|
||||
extensions_update_list = gr.Text(elem_id="extensions_update_list", visible=False).style(container=False)
|
||||
|
||||
info = gr.HTML()
|
||||
extensions_table = gr.HTML(lambda: extension_table())
|
||||
|
||||
apply.click(
|
||||
@ -289,10 +297,10 @@ def create_ui():
|
||||
)
|
||||
|
||||
check.click(
|
||||
fn=check_updates,
|
||||
fn=wrap_gradio_gpu_call(check_updates, extra_outputs=[gr.update()]),
|
||||
_js="extensions_check",
|
||||
inputs=[],
|
||||
outputs=[extensions_table],
|
||||
inputs=[info, extensions_disabled_list],
|
||||
outputs=[extensions_table, info],
|
||||
)
|
||||
|
||||
with gr.TabItem("Available"):
|
||||
|
||||
@ -1,4 +1,7 @@
|
||||
import glob
|
||||
import os.path
|
||||
import urllib.parse
|
||||
from pathlib import Path
|
||||
|
||||
from modules import shared
|
||||
import gradio as gr
|
||||
@ -8,12 +11,31 @@ import html
|
||||
from modules.generation_parameters_copypaste import image_from_url_text
|
||||
|
||||
extra_pages = []
|
||||
allowed_dirs = set()
|
||||
|
||||
|
||||
def register_page(page):
|
||||
"""registers extra networks page for the UI; recommend doing it in on_before_ui() callback for extensions"""
|
||||
|
||||
extra_pages.append(page)
|
||||
allowed_dirs.clear()
|
||||
allowed_dirs.update(set(sum([x.allowed_directories_for_previews() for x in extra_pages], [])))
|
||||
|
||||
|
||||
def add_pages_to_demo(app):
|
||||
def fetch_file(filename: str = ""):
|
||||
from starlette.responses import FileResponse
|
||||
|
||||
if not any([Path(x).resolve() in Path(filename).resolve().parents for x in allowed_dirs]):
|
||||
raise ValueError(f"File cannot be fetched: {filename}. Must be in one of directories registered by extra pages.")
|
||||
|
||||
if os.path.splitext(filename)[1].lower() != ".png":
|
||||
raise ValueError(f"File cannot be fetched: {filename}. Only png.")
|
||||
|
||||
# would profit from returning 304
|
||||
return FileResponse(filename, headers={"Accept-Ranges": "bytes"})
|
||||
|
||||
app.add_api_route("/sd_extra_networks/thumb", fetch_file, methods=["GET"])
|
||||
|
||||
|
||||
class ExtraNetworksPage:
|
||||
@ -26,10 +48,44 @@ class ExtraNetworksPage:
|
||||
def refresh(self):
|
||||
pass
|
||||
|
||||
def link_preview(self, filename):
|
||||
return "./sd_extra_networks/thumb?filename=" + urllib.parse.quote(filename.replace('\\', '/')) + "&mtime=" + str(os.path.getmtime(filename))
|
||||
|
||||
def search_terms_from_path(self, filename, possible_directories=None):
|
||||
abspath = os.path.abspath(filename)
|
||||
|
||||
for parentdir in (possible_directories if possible_directories is not None else self.allowed_directories_for_previews()):
|
||||
parentdir = os.path.abspath(parentdir)
|
||||
if abspath.startswith(parentdir):
|
||||
return abspath[len(parentdir):].replace('\\', '/')
|
||||
|
||||
return ""
|
||||
|
||||
def create_html(self, tabname):
|
||||
view = shared.opts.extra_networks_default_view
|
||||
items_html = ''
|
||||
|
||||
subdirs = {}
|
||||
for parentdir in [os.path.abspath(x) for x in self.allowed_directories_for_previews()]:
|
||||
for x in glob.glob(os.path.join(parentdir, '**/*'), recursive=True):
|
||||
if not os.path.isdir(x):
|
||||
continue
|
||||
|
||||
subdir = os.path.abspath(x)[len(parentdir):].replace("\\", "/")
|
||||
while subdir.startswith("/"):
|
||||
subdir = subdir[1:]
|
||||
|
||||
subdirs[subdir] = 1
|
||||
|
||||
if subdirs:
|
||||
subdirs = {"": 1, **subdirs}
|
||||
|
||||
subdirs_html = "".join([f"""
|
||||
<button class='gr-button gr-button-lg gr-button-secondary{" search-all" if subdir=="" else ""}' onclick='extraNetworksSearchButton("{tabname}_extra_tabs", event)'>
|
||||
{html.escape(subdir if subdir!="" else "all")}
|
||||
</button>
|
||||
""" for subdir in subdirs])
|
||||
|
||||
for item in self.list_items():
|
||||
items_html += self.create_html_for_item(item, tabname)
|
||||
|
||||
@ -38,6 +94,9 @@ class ExtraNetworksPage:
|
||||
items_html = shared.html("extra-networks-no-cards.html").format(dirs=dirs)
|
||||
|
||||
res = f"""
|
||||
<div id='{tabname}_{self.name}_subdirs' class='extra-network-subdirs extra-network-subdirs-{view}'>
|
||||
{subdirs_html}
|
||||
</div>
|
||||
<div id='{tabname}_{self.name}_cards' class='extra-network-{view}'>
|
||||
{items_html}
|
||||
</div>
|
||||
@ -54,14 +113,19 @@ class ExtraNetworksPage:
|
||||
def create_html_for_item(self, item, tabname):
|
||||
preview = item.get("preview", None)
|
||||
|
||||
onclick = item.get("onclick", None)
|
||||
if onclick is None:
|
||||
onclick = '"' + html.escape(f"""return cardClicked({json.dumps(tabname)}, {item["prompt"]}, {"true" if self.allow_negative_prompt else "false"})""") + '"'
|
||||
|
||||
args = {
|
||||
"preview_html": "style='background-image: url(\"" + html.escape(preview) + "\")'" if preview else '',
|
||||
"prompt": item["prompt"],
|
||||
"prompt": item.get("prompt", None),
|
||||
"tabname": json.dumps(tabname),
|
||||
"local_preview": json.dumps(item["local_preview"]),
|
||||
"name": item["name"],
|
||||
"card_clicked": '"' + html.escape(f"""return cardClicked({json.dumps(tabname)}, {item["prompt"]}, {"true" if self.allow_negative_prompt else "false"})""") + '"',
|
||||
"card_clicked": onclick,
|
||||
"save_card_preview": '"' + html.escape(f"""return saveCardPreview(event, {json.dumps(tabname)}, {json.dumps(item["local_preview"])})""") + '"',
|
||||
"search_term": item.get("search_term", ""),
|
||||
}
|
||||
|
||||
return self.card_page.format(**args)
|
||||
@ -143,7 +207,7 @@ def path_is_parent(parent_path, child_path):
|
||||
parent_path = os.path.abspath(parent_path)
|
||||
child_path = os.path.abspath(child_path)
|
||||
|
||||
return os.path.commonpath([parent_path]) == os.path.commonpath([parent_path, child_path])
|
||||
return child_path.startswith(parent_path)
|
||||
|
||||
|
||||
def setup_ui(ui, gallery):
|
||||
@ -173,7 +237,8 @@ def setup_ui(ui, gallery):
|
||||
|
||||
ui.button_save_preview.click(
|
||||
fn=save_preview,
|
||||
_js="function(x, y, z){console.log(x, y, z); return [selected_gallery_index(), y, z]}",
|
||||
_js="function(x, y, z){return [selected_gallery_index(), y, z]}",
|
||||
inputs=[ui.preview_target_filename, gallery, ui.preview_target_filename],
|
||||
outputs=[*ui.pages]
|
||||
)
|
||||
|
||||
|
||||
39
modules/ui_extra_networks_checkpoints.py
Normal file
39
modules/ui_extra_networks_checkpoints.py
Normal file
@ -0,0 +1,39 @@
|
||||
import html
|
||||
import json
|
||||
import os
|
||||
import urllib.parse
|
||||
|
||||
from modules import shared, ui_extra_networks, sd_models
|
||||
|
||||
|
||||
class ExtraNetworksPageCheckpoints(ui_extra_networks.ExtraNetworksPage):
|
||||
def __init__(self):
|
||||
super().__init__('Checkpoints')
|
||||
|
||||
def refresh(self):
|
||||
shared.refresh_checkpoints()
|
||||
|
||||
def list_items(self):
|
||||
checkpoint: sd_models.CheckpointInfo
|
||||
for name, checkpoint in sd_models.checkpoints_list.items():
|
||||
path, ext = os.path.splitext(checkpoint.filename)
|
||||
previews = [path + ".png", path + ".preview.png"]
|
||||
|
||||
preview = None
|
||||
for file in previews:
|
||||
if os.path.isfile(file):
|
||||
preview = self.link_preview(file)
|
||||
break
|
||||
|
||||
yield {
|
||||
"name": checkpoint.name_for_extra,
|
||||
"filename": path,
|
||||
"preview": preview,
|
||||
"search_term": self.search_terms_from_path(checkpoint.filename) + " " + (checkpoint.sha256 or ""),
|
||||
"onclick": '"' + html.escape(f"""return selectCheckpoint({json.dumps(name)})""") + '"',
|
||||
"local_preview": path + ".png",
|
||||
}
|
||||
|
||||
def allowed_directories_for_previews(self):
|
||||
return [v for v in [shared.cmd_opts.ckpt_dir, sd_models.model_path] if v is not None]
|
||||
|
||||
@ -19,13 +19,14 @@ class ExtraNetworksPageHypernetworks(ui_extra_networks.ExtraNetworksPage):
|
||||
preview = None
|
||||
for file in previews:
|
||||
if os.path.isfile(file):
|
||||
preview = "./file=" + file.replace('\\', '/') + "?mtime=" + str(os.path.getmtime(file))
|
||||
preview = self.link_preview(file)
|
||||
break
|
||||
|
||||
yield {
|
||||
"name": name,
|
||||
"filename": path,
|
||||
"preview": preview,
|
||||
"search_term": self.search_terms_from_path(path),
|
||||
"prompt": json.dumps(f"<hypernet:{name}:") + " + opts.extra_networks_default_multiplier + " + json.dumps(">"),
|
||||
"local_preview": path + ".png",
|
||||
}
|
||||
|
||||
@ -19,12 +19,13 @@ class ExtraNetworksPageTextualInversion(ui_extra_networks.ExtraNetworksPage):
|
||||
|
||||
preview = None
|
||||
if os.path.isfile(preview_file):
|
||||
preview = "./file=" + preview_file.replace('\\', '/') + "?mtime=" + str(os.path.getmtime(preview_file))
|
||||
preview = self.link_preview(preview_file)
|
||||
|
||||
yield {
|
||||
"name": embedding.name,
|
||||
"filename": embedding.filename,
|
||||
"preview": preview,
|
||||
"search_term": self.search_terms_from_path(embedding.filename),
|
||||
"prompt": json.dumps(embedding.name),
|
||||
"local_preview": path + ".preview.png",
|
||||
}
|
||||
|
||||
@ -383,6 +383,15 @@ class Script(scripts.Script):
|
||||
y_type.change(fn=select_axis, inputs=[y_type], outputs=[fill_y_button])
|
||||
z_type.change(fn=select_axis, inputs=[z_type], outputs=[fill_z_button])
|
||||
|
||||
self.infotext_fields = (
|
||||
(x_type, "X Type"),
|
||||
(x_values, "X Values"),
|
||||
(y_type, "Y Type"),
|
||||
(y_values, "Y Values"),
|
||||
(z_type, "Z Type"),
|
||||
(z_values, "Z Values"),
|
||||
)
|
||||
|
||||
return [x_type, x_values, y_type, y_values, z_type, z_values, draw_legend, include_lone_images, include_sub_grids, no_fixed_seeds]
|
||||
|
||||
def run(self, p, x_type, x_values, y_type, y_values, z_type, z_values, draw_legend, include_lone_images, include_sub_grids, no_fixed_seeds):
|
||||
@ -542,6 +551,7 @@ class Script(scripts.Script):
|
||||
|
||||
if grid_infotext[0] is None:
|
||||
pc.extra_generation_params = copy(pc.extra_generation_params)
|
||||
pc.extra_generation_params['Script'] = self.title()
|
||||
|
||||
if x_opt.label != 'Nothing':
|
||||
pc.extra_generation_params["X Type"] = x_opt.label
|
||||
|
||||
46
style.css
46
style.css
@ -74,7 +74,12 @@
|
||||
#txt2img_gallery img, #img2img_gallery img{
|
||||
object-fit: scale-down;
|
||||
}
|
||||
|
||||
#txt2img_actions_column, #img2img_actions_column {
|
||||
margin: 0.35rem 0.75rem 0.35rem 0;
|
||||
}
|
||||
#script_list {
|
||||
padding: .625rem .75rem 0 .625rem;
|
||||
}
|
||||
.justify-center.overflow-x-scroll {
|
||||
justify-content: left;
|
||||
}
|
||||
@ -126,6 +131,7 @@
|
||||
|
||||
#txt2img_actions_column, #img2img_actions_column{
|
||||
gap: 0;
|
||||
margin-right: .75rem;
|
||||
}
|
||||
|
||||
#txt2img_tools, #img2img_tools{
|
||||
@ -150,6 +156,7 @@
|
||||
|
||||
#txt2img_styles_row, #img2img_styles_row{
|
||||
gap: 0.25em;
|
||||
margin-top: 0.3em;
|
||||
}
|
||||
|
||||
#txt2img_styles_row > button, #img2img_styles_row > button{
|
||||
@ -311,11 +318,11 @@ input[type="range"]{
|
||||
.min-h-\[6rem\] { min-height: unset !important; }
|
||||
|
||||
.progressDiv{
|
||||
position: absolute;
|
||||
position: relative;
|
||||
height: 20px;
|
||||
top: -20px;
|
||||
background: #b4c0cc;
|
||||
border-radius: 3px !important;
|
||||
margin-bottom: -3px;
|
||||
}
|
||||
|
||||
.dark .progressDiv{
|
||||
@ -535,7 +542,7 @@ input[type="range"]{
|
||||
}
|
||||
|
||||
#quicksettings {
|
||||
gap: 0.4em;
|
||||
width: fit-content;
|
||||
}
|
||||
|
||||
#quicksettings > div, #quicksettings > fieldset{
|
||||
@ -545,6 +552,7 @@ input[type="range"]{
|
||||
border: none;
|
||||
box-shadow: none;
|
||||
background: none;
|
||||
margin-right: 10px;
|
||||
}
|
||||
|
||||
#quicksettings > div > div > div > label > span {
|
||||
@ -567,7 +575,7 @@ canvas[key="mask"] {
|
||||
right: 0.5em;
|
||||
top: -0.6em;
|
||||
z-index: 400;
|
||||
width: 8em;
|
||||
width: 6em;
|
||||
}
|
||||
#quicksettings .gr-box > div > div > input.gr-text-input {
|
||||
top: -1.12em;
|
||||
@ -665,11 +673,27 @@ canvas[key="mask"] {
|
||||
|
||||
#quicksettings .gr-button-tool{
|
||||
margin: 0;
|
||||
border-color: unset;
|
||||
background-color: unset;
|
||||
}
|
||||
|
||||
|
||||
#modelmerger_interp_description>p {
|
||||
margin: 0!important;
|
||||
text-align: center;
|
||||
}
|
||||
#modelmerger_interp_description {
|
||||
margin: 0.35rem 0.75rem 1.23rem;
|
||||
}
|
||||
#img2img_settings > div.gr-form, #txt2img_settings > div.gr-form {
|
||||
padding-top: 0.9em;
|
||||
padding-bottom: 0.9em;
|
||||
}
|
||||
#txt2img_settings {
|
||||
padding-top: 1.16em;
|
||||
padding-bottom: 0.9em;
|
||||
}
|
||||
#img2img_settings {
|
||||
padding-bottom: 0.9em;
|
||||
}
|
||||
|
||||
#img2img_settings div.gr-form .gr-form, #txt2img_settings div.gr-form .gr-form, #train_tabs div.gr-form .gr-form{
|
||||
@ -741,6 +765,7 @@ footer {
|
||||
|
||||
.dark .gr-compact{
|
||||
background-color: rgb(31 41 55 / var(--tw-bg-opacity));
|
||||
margin-left: 0;
|
||||
}
|
||||
|
||||
.gr-compact{
|
||||
@ -782,7 +807,13 @@ footer {
|
||||
margin: 0.3em;
|
||||
}
|
||||
|
||||
.extra-network-subdirs{
|
||||
padding: 0.2em 0.35em;
|
||||
}
|
||||
|
||||
.extra-network-subdirs button{
|
||||
margin: 0 0.15em;
|
||||
}
|
||||
|
||||
#txt2img_extra_networks .search, #img2img_extra_networks .search{
|
||||
display: inline-block;
|
||||
@ -925,3 +956,6 @@ footer {
|
||||
color: red;
|
||||
}
|
||||
|
||||
[id*='_prompt_container'] > div {
|
||||
margin: 0!important;
|
||||
}
|
||||
|
||||
@ -10,7 +10,7 @@ then
|
||||
fi
|
||||
|
||||
export install_dir="$HOME"
|
||||
export COMMANDLINE_ARGS="--skip-torch-cuda-test --upcast-sampling --use-cpu interrogate"
|
||||
export COMMANDLINE_ARGS="--skip-torch-cuda-test --upcast-sampling --no-half-vae --use-cpu interrogate"
|
||||
export TORCH_COMMAND="pip install torch==1.12.1 torchvision==0.13.1"
|
||||
export K_DIFFUSION_REPO="https://github.com/brkirch/k-diffusion.git"
|
||||
export K_DIFFUSION_COMMIT_HASH="51c9778f269cedb55a4d88c79c0246d35bdadb71"
|
||||
|
||||
6
webui.py
6
webui.py
@ -12,7 +12,7 @@ from packaging import version
|
||||
import logging
|
||||
logging.getLogger("xformers").addFilter(lambda record: 'A matching Triton is not available' not in record.getMessage())
|
||||
|
||||
from modules import import_hook, errors, extra_networks
|
||||
from modules import import_hook, errors, extra_networks, ui_extra_networks_checkpoints
|
||||
from modules import extra_networks_hypernet, ui_extra_networks_hypernets, ui_extra_networks_textual_inversion
|
||||
from modules.call_queue import wrap_queued_call, queue_lock, wrap_gradio_gpu_call
|
||||
|
||||
@ -119,6 +119,7 @@ def initialize():
|
||||
ui_extra_networks.intialize()
|
||||
ui_extra_networks.register_page(ui_extra_networks_textual_inversion.ExtraNetworksPageTextualInversion())
|
||||
ui_extra_networks.register_page(ui_extra_networks_hypernets.ExtraNetworksPageHypernetworks())
|
||||
ui_extra_networks.register_page(ui_extra_networks_checkpoints.ExtraNetworksPageCheckpoints())
|
||||
|
||||
extra_networks.initialize()
|
||||
extra_networks.register_extra_network(extra_networks_hypernet.ExtraNetworkHypernet())
|
||||
@ -227,6 +228,8 @@ def webui():
|
||||
if launch_api:
|
||||
create_api(app)
|
||||
|
||||
ui_extra_networks.add_pages_to_demo(app)
|
||||
|
||||
modules.script_callbacks.app_started_callback(shared.demo, app)
|
||||
|
||||
wait_on_server(shared.demo)
|
||||
@ -254,6 +257,7 @@ def webui():
|
||||
ui_extra_networks.intialize()
|
||||
ui_extra_networks.register_page(ui_extra_networks_textual_inversion.ExtraNetworksPageTextualInversion())
|
||||
ui_extra_networks.register_page(ui_extra_networks_hypernets.ExtraNetworksPageHypernetworks())
|
||||
ui_extra_networks.register_page(ui_extra_networks_checkpoints.ExtraNetworksPageCheckpoints())
|
||||
|
||||
extra_networks.initialize()
|
||||
extra_networks.register_extra_network(extra_networks_hypernet.ExtraNetworkHypernet())
|
||||
|
||||
Loading…
Reference in New Issue
Block a user