nodes/mcmonkeyprojects-sd-dynamic-thresholding
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

add 'power up' and matching input val

Browse Source
This commit is contained in:
Alex "mcmonkey" Goodwin
2023-02-10 00:55:07 -08:00
parent b9c330db7d
commit 36205bfbf8
1 changed files with 11 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/dynamic_thresholding.py
+11 -6
View File
@@ -17,7 +17,7 @@ import math
from modules import scripts, sd_samplers, sd_samplers_kdiffusion, sd_samplers_common
######################### Data values #########################
VALID_MODES = ["Constant", "Linear Down", "Cosine Down", "Half Cosine Down", "Linear Up", "Cosine Up", "Half Cosine Up"]
VALID_MODES = ["Constant", "Linear Down", "Cosine Down", "Half Cosine Down", "Linear Up", "Cosine Up", "Half Cosine Up", "Power Up"]
######################### Script class entrypoint #########################
class Script(scripts.Script):
@@ -41,16 +41,17 @@ class Script(scripts.Script):
mimic_scale_min = gr.Slider(minimum=0.0, maximum=30.0, step=0.5, label="Minimum value of the Mimic Scale Scheduler")
cfg_mode = gr.Dropdown(VALID_MODES, value="Constant", label="CFG Scale Scheduler")
cfg_scale_min = gr.Slider(minimum=0.0, maximum=30.0, step=0.5, label="Minimum value of the CFG Scale Scheduler")
powerscale_power = gr.Slider(minimum=0.0, maximum=15.0, step=0.5, value=4.0, label="Power Scheduler Value")
enabled.change(
fn=lambda x: {"visible": x, "__type__": "update"},
inputs=[enabled],
outputs=[accordion],
show_progress = False)
return [enabled, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min]
return [enabled, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, powerscale_power]
last_id = 0
def process_batch(self, p, enabled, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, batch_number, prompts, seeds, subseeds):
def process_batch(self, p, enabled, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, powerscale_power, batch_number, prompts, seeds, subseeds):
enabled = p.dynthres_enabled if hasattr(p, 'dynthres_enabled') else enabled
if not enabled:
return
@@ -63,6 +64,7 @@ class Script(scripts.Script):
cfg_mode = p.dynthres_cfg_mode if hasattr(p, 'dynthres_cfg_mode') else cfg_mode
cfg_scale_min = p.dynthres_cfg_scale_min if hasattr(p, 'dynthres_cfg_scale_min') else cfg_scale_min
experiment_mode = p.dynthres_experiment_mode if hasattr(p, 'dynthres_experiment_mode') else 0
power_val = p.dynthres_power_val if hasattr(p, 'dynthres_power_val') else powerscale_power
p.extra_generation_params["Dynamic thresholding enabled"] = True
p.extra_generation_params["Mimic scale"] = mimic_scale
p.extra_generation_params["Threshold percentile"] = threshold_percentile
@@ -81,7 +83,7 @@ class Script(scripts.Script):
sampler = sd_samplers.all_samplers_map[p.sampler_name]
def newConstructor(model):
result = sampler.constructor(model)
cfg = CustomCFGDenoiser(result.model_wrap_cfg.inner_model, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, experiment_mode, p.steps)
cfg = CustomCFGDenoiser(result.model_wrap_cfg.inner_model, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, power_val, experiment_mode, p.steps)
result.model_wrap_cfg = cfg
return result
newSampler = sd_samplers_common.SamplerData(fixed_sampler_name, newConstructor, sampler.aliases, sampler.options)
@@ -91,7 +93,7 @@ class Script(scripts.Script):
p.fixed_sampler_name = fixed_sampler_name
sd_samplers.all_samplers_map[fixed_sampler_name] = newSampler
def postprocess_batch(self, p, enabled, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, batch_number, images):
def postprocess_batch(self, p, enabled, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, powerscale_power, batch_number, images):
if not enabled or not hasattr(p, 'orig_sampler_name'):
return
p.sampler_name = p.orig_sampler_name
@@ -102,7 +104,7 @@ class Script(scripts.Script):
######################### Implementation logic #########################
class CustomCFGDenoiser(sd_samplers_kdiffusion.CFGDenoiser):
def __init__(self, model, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, experiment_mode, maxSteps):
def __init__(self, model, mimic_scale, threshold_percentile, mimic_mode, mimic_scale_min, cfg_mode, cfg_scale_min, power_val, experiment_mode, maxSteps):
super().__init__(model)
self.mimic_scale = mimic_scale
self.threshold_percentile = threshold_percentile
@@ -112,6 +114,7 @@ class CustomCFGDenoiser(sd_samplers_kdiffusion.CFGDenoiser):
self.cfg_scale_min = cfg_scale_min
self.mimic_scale_min = mimic_scale_min
self.experiment_mode = experiment_mode
self.power_val = power_val
def combine_denoised(self, x_out, conds_list, uncond, cond_scale):
denoised_uncond = x_out[-uncond.shape[0]:]
@@ -134,6 +137,8 @@ class CustomCFGDenoiser(sd_samplers_kdiffusion.CFGDenoiser):
scale *= 1.0 - math.cos((self.step / max))
elif mode == "Cosine Up":
scale *= 1.0 - math.cos((self.step / max) * 1.5707)
elif mode == "Power Up":
scale *= math.pow(self.step / max, self.power_val)
scale += min
return scale