File size: 2,354 Bytes
0070fce
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
"""
Credit: ComfyUI
https://github.com/comfyanonymous/ComfyUI/blob/v0.3.7/comfy_extras/nodes_model_advanced.py#L257
"""

import gradio as gr
import torch
from modules import scripts
from modules.infotext_utils import PasteField
from modules.shared import opts


class ScriptRescaleCFG(scripts.ScriptBuiltinUI):
    section = "cfg"
    create_group = False

    def title(self):
        return "RescaleCFG"

    def show(self, is_img2img):
        return scripts.AlwaysVisible if opts.show_rescale_cfg else None

    def ui(self, is_img2img):
        cfg = gr.Slider(
            value=0.0,
            minimum=0.0,
            maximum=1.0,
            step=0.05,
            label="Rescale CFG",
            elem_id=f"{'img2img' if is_img2img else 'txt2img'}_rescale_cfg_scale",
            scale=4,
        )

        self.infotext_fields = [PasteField(cfg, "Rescale CFG", api="rescale_cfg")]

        return [cfg]

    def after_extra_networks_activate(self, p, cfg, *args, **kwargs):
        if opts.show_rescale_cfg and cfg > 0.0:
            p.extra_generation_params.update({"Rescale CFG": cfg})

    def process_before_every_sampling(self, p, cfg, *args, **kwargs):
        if not opts.show_rescale_cfg or cfg < 0.05:
            return

        @torch.inference_mode()
        def rescale_cfg(args):
            cond = args["cond"]
            uncond = args["uncond"]
            cond_scale = args["cond_scale"]
            sigma = args["sigma"]
            sigma = sigma.view(sigma.shape[:1] + (1,) * (cond.ndim - 1))
            x_orig = args["input"]

            x = x_orig / (sigma * sigma + 1.0)
            cond = ((x - (x_orig - cond)) * (sigma**2 + 1.0) ** 0.5) / (sigma)
            uncond = ((x - (x_orig - uncond)) * (sigma**2 + 1.0) ** 0.5) / (sigma)

            x_cfg = uncond + cond_scale * (cond - uncond)
            ro_pos = torch.std(cond, dim=(1, 2, 3), keepdim=True)
            ro_cfg = torch.std(x_cfg, dim=(1, 2, 3), keepdim=True)

            x_rescaled = x_cfg * (ro_pos / ro_cfg)
            x_final = cfg * x_rescaled + (1.0 - cfg) * x_cfg

            return x_orig - (x - x_final * sigma / (sigma * sigma + 1.0) ** 0.5)

        unet = p.sd_model.forge_objects.unet.clone()
        unet.set_model_sampler_cfg_function(rescale_cfg)
        p.sd_model.forge_objects.unet = unet

        print(f"rescale_cfg = {cfg}")