hoelee/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-08-04 03:10:21 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Added slider for detail preservation strength, removed largely needless offset parameter, changed labels in UI and for saving to/pasting data from PNG files.

Browse Source
This commit is contained in:
CodeHatchling
2023-11-28 22:35:07 -07:00
parent debf836fcc
commit c5c7fa06aa
8 changed files with 32 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
modules/sd_samplers_cfg_denoiser.py
View File

@@ -45,7 +45,7 @@ class CFGDenoiser(torch.nn.Module):
self.nmask = None
self.mask_blend_power = 1
self.mask_blend_scale = 1
self.mask_blend_offset = 0
self.inpaint_detail_preservation = 16
self.init_latent = None
self.steps = None
"""number of steps as specified by user in UI"""
@@ -105,14 +105,13 @@ class CFGDenoiser(torch.nn.Module):
# Record the original latent vector magnitudes.
# We bring them to a power so that larger magnitudes are favored over smaller ones.
# 64-bit operations are used here to allow large exponents.
detail_preservation = 32
a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64) ** detail_preservation
b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64) ** detail_preservation
a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64) ** self.inpaint_detail_preservation
b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64) ** self.inpaint_detail_preservation
one_minus_t = 1 - t
# Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1).
interp_magnitude = (a_magnitude * one_minus_t + b_magnitude * t) ** (1 / detail_preservation)
interp_magnitude = (a_magnitude * one_minus_t + b_magnitude * t) ** (1 / self.inpaint_detail_preservation)
# Linearly interpolate the image vectors.
image_interp = a * one_minus_t + b * t
@@ -142,7 +141,7 @@ class CFGDenoiser(torch.nn.Module):
NOTE: "mask" is not used
"""
return torch.pow(nmask, (_sigma ** self.mask_blend_power) * self.mask_blend_scale + self.mask_blend_offset)
return torch.pow(nmask, (_sigma ** self.mask_blend_power) * self.mask_blend_scale)
if state.interrupted or state.skipped:
raise sd_samplers_common.InterruptedException