stable-diffusion-webui/scripts/xy_grid.py

391 lines
15 KiB
Python
Raw Normal View History

2022-09-03 19:32:45 +03:00
from collections import namedtuple
from copy import copy
2022-10-06 13:55:21 +03:00
from itertools import permutations, chain
2022-09-03 19:32:45 +03:00
import random
2022-10-06 13:55:21 +03:00
import csv
from io import StringIO
from PIL import Image
import numpy as np
2022-09-03 19:32:45 +03:00
import modules.scripts as scripts
import gradio as gr
2022-10-11 14:53:02 +03:00
from modules import images
from modules.hypernetworks import hypernetwork
2022-10-09 03:13:13 +03:00
from modules.processing import process_images, Processed, get_correct_sampler
2022-09-03 19:32:45 +03:00
from modules.shared import opts, cmd_opts, state
import modules.shared as shared
2022-09-03 19:32:45 +03:00
import modules.sd_samplers
import modules.sd_models
import re
2022-09-03 19:32:45 +03:00
def apply_field(field):
def fun(p, x, xs):
setattr(p, field, x)
return fun
def apply_prompt(p, x, xs):
if xs[0] not in p.prompt and xs[0] not in p.negative_prompt:
raise RuntimeError(f"Prompt S/R did not find {xs[0]} in prompt or negative prompt.")
2022-09-03 19:32:45 +03:00
p.prompt = p.prompt.replace(xs[0], x)
p.negative_prompt = p.negative_prompt.replace(xs[0], x)
2022-09-03 19:32:45 +03:00
2022-10-04 09:18:00 +03:00
def apply_order(p, x, xs):
token_order = []
2022-10-04 09:18:00 +03:00
# Initally grab the tokens from the prompt, so they can be replaced in order of earliest seen
for token in x:
token_order.append((p.prompt.find(token), token))
token_order.sort(key=lambda t: t[0])
2022-10-04 08:07:36 +03:00
prompt_parts = []
# Split the prompt up, taking out the tokens
for _, token in token_order:
n = p.prompt.find(token)
prompt_parts.append(p.prompt[0:n])
p.prompt = p.prompt[n + len(token):]
# Rebuild the prompt with the tokens in the order we want
prompt_tmp = ""
for idx, part in enumerate(prompt_parts):
prompt_tmp += part
prompt_tmp += x[idx]
p.prompt = prompt_tmp + p.prompt
2022-09-03 19:32:45 +03:00
2022-10-09 03:13:13 +03:00
def build_samplers_dict(p):
samplers_dict = {}
for i, sampler in enumerate(get_correct_sampler(p)):
samplers_dict[sampler.name.lower()] = i
for alias in sampler.aliases:
samplers_dict[alias.lower()] = i
return samplers_dict
2022-09-03 19:32:45 +03:00
def apply_sampler(p, x, xs):
2022-10-09 03:13:13 +03:00
sampler_index = build_samplers_dict(p).get(x.lower(), None)
2022-09-03 19:32:45 +03:00
if sampler_index is None:
raise RuntimeError(f"Unknown sampler: {x}")
p.sampler_index = sampler_index
2022-10-09 20:20:35 +03:00
def confirm_samplers(p, xs):
samplers_dict = build_samplers_dict(p)
for x in xs:
if x.lower() not in samplers_dict.keys():
raise RuntimeError(f"Unknown sampler: {x}")
def apply_checkpoint(p, x, xs):
2022-09-29 00:31:53 +03:00
info = modules.sd_models.get_closet_checkpoint_match(x)
2022-10-09 20:20:35 +03:00
if info is None:
raise RuntimeError(f"Unknown checkpoint: {x}")
modules.sd_models.reload_model_weights(shared.sd_model, info)
2022-10-09 20:20:35 +03:00
def confirm_checkpoints(p, xs):
for x in xs:
if modules.sd_models.get_closet_checkpoint_match(x) is None:
raise RuntimeError(f"Unknown checkpoint: {x}")
2022-10-07 10:17:52 +03:00
def apply_hypernetwork(p, x, xs):
if x.lower() in ["", "none"]:
name = None
else:
name = hypernetwork.find_closest_hypernetwork_name(x)
if not name:
raise RuntimeError(f"Unknown hypernetwork: {x}")
hypernetwork.load_hypernetwork(name)
2022-10-07 10:17:52 +03:00
2022-10-13 20:12:37 +03:00
def apply_hypernetwork_strength(p, x, xs):
hypernetwork.apply_strength(x)
2022-10-09 20:20:35 +03:00
def confirm_hypernetworks(p, xs):
for x in xs:
if x.lower() in ["", "none"]:
continue
if not hypernetwork.find_closest_hypernetwork_name(x):
raise RuntimeError(f"Unknown hypernetwork: {x}")
2022-10-09 18:58:55 +03:00
def apply_clip_skip(p, x, xs):
opts.data["CLIP_stop_at_last_layers"] = x
2022-10-09 18:58:55 +03:00
2022-09-03 19:32:45 +03:00
def format_value_add_label(p, opt, x):
if type(x) == float:
x = round(x, 8)
2022-09-03 19:32:45 +03:00
return f"{opt.label}: {x}"
def format_value(p, opt, x):
if type(x) == float:
x = round(x, 8)
2022-09-03 19:32:45 +03:00
return x
2022-10-04 09:18:00 +03:00
def format_value_join_list(p, opt, x):
return ", ".join(x)
def do_nothing(p, x, xs):
pass
2022-10-04 09:18:00 +03:00
def format_nothing(p, opt, x):
return ""
2022-09-03 19:32:45 +03:00
2022-10-04 09:18:00 +03:00
def str_permutations(x):
"""dummy function for specifying it in AxisOption's type when you want to get a list of permutations"""
return x
2022-10-09 20:20:35 +03:00
AxisOption = namedtuple("AxisOption", ["label", "type", "apply", "format_value", "confirm"])
AxisOptionImg2Img = namedtuple("AxisOptionImg2Img", ["label", "type", "apply", "format_value", "confirm"])
2022-09-03 19:32:45 +03:00
axis_options = [
2022-10-09 20:20:35 +03:00
AxisOption("Nothing", str, do_nothing, format_nothing, None),
AxisOption("Seed", int, apply_field("seed"), format_value_add_label, None),
AxisOption("Var. seed", int, apply_field("subseed"), format_value_add_label, None),
AxisOption("Var. strength", float, apply_field("subseed_strength"), format_value_add_label, None),
AxisOption("Steps", int, apply_field("steps"), format_value_add_label, None),
AxisOption("CFG Scale", float, apply_field("cfg_scale"), format_value_add_label, None),
AxisOption("Prompt S/R", str, apply_prompt, format_value, None),
AxisOption("Prompt order", str_permutations, apply_order, format_value_join_list, None),
AxisOption("Sampler", str, apply_sampler, format_value, confirm_samplers),
AxisOption("Checkpoint name", str, apply_checkpoint, format_value, confirm_checkpoints),
AxisOption("Hypernetwork", str, apply_hypernetwork, format_value, confirm_hypernetworks),
2022-10-13 20:12:37 +03:00
AxisOption("Hypernet str.", float, apply_hypernetwork_strength, format_value_add_label, None),
2022-10-09 20:20:35 +03:00
AxisOption("Sigma Churn", float, apply_field("s_churn"), format_value_add_label, None),
AxisOption("Sigma min", float, apply_field("s_tmin"), format_value_add_label, None),
AxisOption("Sigma max", float, apply_field("s_tmax"), format_value_add_label, None),
AxisOption("Sigma noise", float, apply_field("s_noise"), format_value_add_label, None),
AxisOption("Eta", float, apply_field("eta"), format_value_add_label, None),
AxisOption("Clip skip", int, apply_clip_skip, format_value_add_label, None),
AxisOptionImg2Img("Denoising", float, apply_field("denoising_strength"), format_value_add_label, None), # as it is now all AxisOptionImg2Img items must go after AxisOption ones
2022-09-03 19:32:45 +03:00
]
def draw_xy_grid(p, xs, ys, x_labels, y_labels, cell, draw_legend, include_lone_images):
ver_texts = [[images.GridAnnotation(y)] for y in y_labels]
hor_texts = [[images.GridAnnotation(x)] for x in x_labels]
2022-09-03 19:32:45 +03:00
2022-10-13 02:12:12 +03:00
# Temporary list of all the images that are generated to be populated into the grid.
# Will be filled with empty images for any individual step that fails to process properly
image_cache = []
processed_result = None
cell_mode = "P"
cell_size = (1,1)
2022-09-03 19:32:45 +03:00
state.job_count = len(xs) * len(ys) * p.n_iter
2022-09-03 19:32:45 +03:00
for iy, y in enumerate(ys):
for ix, x in enumerate(xs):
state.job = f"{ix + iy * len(xs) + 1} out of {len(xs) * len(ys)}"
2022-09-03 19:32:45 +03:00
2022-10-13 02:12:12 +03:00
processed:Processed = cell(x, y)
try:
2022-10-13 02:12:12 +03:00
# this dereference will throw an exception if the image was not processed
# (this happens in cases such as if the user stops the process from the UI)
processed_image = processed.images[0]
if processed_result is None:
# Use our first valid processed result as a template container to hold our full results
processed_result = copy(processed)
cell_mode = processed_image.mode
cell_size = processed_image.size
processed_result.images = [Image.new(cell_mode, cell_size)]
image_cache.append(processed_image)
if include_lone_images:
processed_result.images.append(processed_image)
processed_result.all_prompts.append(processed.prompt)
processed_result.all_seeds.append(processed.seed)
processed_result.infotexts.append(processed.infotexts[0])
except:
2022-10-13 02:12:12 +03:00
image_cache.append(Image.new(cell_mode, cell_size))
if not processed_result:
print("Unexpected error: draw_xy_grid failed to return even a single processed image")
return Processed()
2022-09-03 19:32:45 +03:00
2022-10-13 02:12:12 +03:00
grid = images.image_grid(image_cache, rows=len(ys))
2022-09-14 15:01:16 +03:00
if draw_legend:
2022-10-13 02:12:12 +03:00
grid = images.draw_grid_annotations(grid, cell_size[0], cell_size[1], hor_texts, ver_texts)
2022-09-03 19:32:45 +03:00
2022-10-13 02:12:12 +03:00
processed_result.images[0] = grid
2022-09-03 19:32:45 +03:00
2022-10-13 02:12:12 +03:00
return processed_result
2022-09-03 19:32:45 +03:00
re_range = re.compile(r"\s*([+-]?\s*\d+)\s*-\s*([+-]?\s*\d+)(?:\s*\(([+-]\d+)\s*\))?\s*")
re_range_float = re.compile(r"\s*([+-]?\s*\d+(?:.\d*)?)\s*-\s*([+-]?\s*\d+(?:.\d*)?)(?:\s*\(([+-]\d+(?:.\d*)?)\s*\))?\s*")
2022-09-14 14:56:26 +03:00
re_range_count = re.compile(r"\s*([+-]?\s*\d+)\s*-\s*([+-]?\s*\d+)(?:\s*\[(\d+)\s*\])?\s*")
re_range_count_float = re.compile(r"\s*([+-]?\s*\d+(?:.\d*)?)\s*-\s*([+-]?\s*\d+(?:.\d*)?)(?:\s*\[(\d+(?:.\d*)?)\s*\])?\s*")
2022-09-03 19:32:45 +03:00
class Script(scripts.Script):
def title(self):
return "X/Y plot"
def ui(self, is_img2img):
current_axis_options = [x for x in axis_options if type(x) == AxisOption or type(x) == AxisOptionImg2Img and is_img2img]
with gr.Row():
x_type = gr.Dropdown(label="X type", choices=[x.label for x in current_axis_options], value=current_axis_options[1].label, visible=False, type="index", elem_id="x_type")
2022-09-03 19:32:45 +03:00
x_values = gr.Textbox(label="X values", visible=False, lines=1)
with gr.Row():
y_type = gr.Dropdown(label="Y type", choices=[x.label for x in current_axis_options], value=current_axis_options[0].label, visible=False, type="index", elem_id="y_type")
2022-09-03 19:32:45 +03:00
y_values = gr.Textbox(label="Y values", visible=False, lines=1)
2022-09-14 15:01:16 +03:00
draw_legend = gr.Checkbox(label='Draw legend', value=True)
2022-10-13 20:12:37 +03:00
include_lone_images = gr.Checkbox(label='Include Separate Images', value=False)
no_fixed_seeds = gr.Checkbox(label='Keep -1 for seeds', value=False)
2022-09-03 19:32:45 +03:00
return [x_type, x_values, y_type, y_values, draw_legend, include_lone_images, no_fixed_seeds]
def run(self, p, x_type, x_values, y_type, y_values, draw_legend, include_lone_images, no_fixed_seeds):
if not no_fixed_seeds:
modules.processing.fix_seed(p)
if not opts.return_grid:
p.batch_size = 1
CLIP_stop_at_last_layers = opts.CLIP_stop_at_last_layers
2022-09-03 19:32:45 +03:00
def process_axis(opt, vals):
if opt.label == 'Nothing':
return [0]
valslist = [x.strip() for x in chain.from_iterable(csv.reader(StringIO(vals)))]
2022-09-03 19:32:45 +03:00
if opt.type == int:
valslist_ext = []
for val in valslist:
m = re_range.fullmatch(val)
2022-09-14 14:56:26 +03:00
mc = re_range_count.fullmatch(val)
if m is not None:
start = int(m.group(1))
end = int(m.group(2))+1
step = int(m.group(3)) if m.group(3) is not None else 1
2022-09-03 19:32:45 +03:00
valslist_ext += list(range(start, end, step))
2022-09-14 14:56:26 +03:00
elif mc is not None:
start = int(mc.group(1))
end = int(mc.group(2))
num = int(mc.group(3)) if mc.group(3) is not None else 1
valslist_ext += [int(x) for x in np.linspace(start=start, stop=end, num=num).tolist()]
2022-09-03 19:32:45 +03:00
else:
valslist_ext.append(val)
valslist = valslist_ext
elif opt.type == float:
valslist_ext = []
for val in valslist:
m = re_range_float.fullmatch(val)
2022-09-14 14:56:26 +03:00
mc = re_range_count_float.fullmatch(val)
if m is not None:
start = float(m.group(1))
end = float(m.group(2))
step = float(m.group(3)) if m.group(3) is not None else 1
valslist_ext += np.arange(start, end + step, step).tolist()
2022-09-14 14:56:26 +03:00
elif mc is not None:
start = float(mc.group(1))
end = float(mc.group(2))
num = int(mc.group(3)) if mc.group(3) is not None else 1
valslist_ext += np.linspace(start=start, stop=end, num=num).tolist()
else:
valslist_ext.append(val)
valslist = valslist_ext
2022-10-04 09:18:00 +03:00
elif opt.type == str_permutations:
valslist = list(permutations(valslist))
2022-09-03 19:32:45 +03:00
valslist = [opt.type(x) for x in valslist]
2022-10-09 20:20:35 +03:00
# Confirm options are valid before starting
2022-10-09 20:20:35 +03:00
if opt.confirm:
opt.confirm(p, valslist)
2022-09-03 19:32:45 +03:00
return valslist
x_opt = axis_options[x_type]
xs = process_axis(x_opt, x_values)
y_opt = axis_options[y_type]
ys = process_axis(y_opt, y_values)
def fix_axis_seeds(axis_opt, axis_list):
2022-10-13 02:35:36 +03:00
if axis_opt.label in ['Seed','Var. seed']:
return [int(random.randrange(4294967294)) if val is None or val == '' or val == -1 else val for val in axis_list]
else:
return axis_list
if not no_fixed_seeds:
xs = fix_axis_seeds(x_opt, xs)
ys = fix_axis_seeds(y_opt, ys)
if x_opt.label == 'Steps':
total_steps = sum(xs) * len(ys)
elif y_opt.label == 'Steps':
total_steps = sum(ys) * len(xs)
else:
total_steps = p.steps * len(xs) * len(ys)
print(f"X/Y plot will create {len(xs) * len(ys) * p.n_iter} images on a {len(xs)}x{len(ys)} grid. (Total steps to process: {total_steps * p.n_iter})")
shared.total_tqdm.updateTotal(total_steps * p.n_iter)
2022-09-03 19:32:45 +03:00
def cell(x, y):
pc = copy(p)
x_opt.apply(pc, x, xs)
y_opt.apply(pc, y, ys)
return process_images(pc)
processed = draw_xy_grid(
p,
2022-09-03 19:32:45 +03:00
xs=xs,
ys=ys,
x_labels=[x_opt.format_value(p, x_opt, x) for x in xs],
y_labels=[y_opt.format_value(p, y_opt, y) for y in ys],
2022-09-14 15:01:16 +03:00
cell=cell,
draw_legend=draw_legend,
include_lone_images=include_lone_images
2022-09-03 19:32:45 +03:00
)
2022-09-04 03:38:24 +03:00
if opts.grid_save:
images.save_image(processed.images[0], p.outpath_grids, "xy_grid", prompt=p.prompt, seed=processed.seed, grid=True, p=p)
2022-09-03 19:32:45 +03:00
# restore checkpoint in case it was changed by axes
modules.sd_models.reload_model_weights(shared.sd_model)
2022-10-09 14:33:22 +03:00
hypernetwork.load_hypernetwork(opts.sd_hypernetwork)
2022-10-13 20:12:37 +03:00
hypernetwork.apply_strength()
2022-10-07 10:17:52 +03:00
opts.data["CLIP_stop_at_last_layers"] = CLIP_stop_at_last_layers
2022-10-09 18:58:55 +03:00
2022-09-03 19:32:45 +03:00
return processed