nodes/Clybius-ComfyUI-Extra-Samplers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Add image weighting for ImageAssistedCFGGuider.

Browse Source
This commit is contained in:
Clybius
2024-04-06 16:03:07 -05:00
parent 39ded6996b
commit ed17033a5a
1 changed files with 14 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
nodes.py
+14 -5
View File
@@ -555,10 +555,11 @@ class GeometricCFGGuider:
return (guider,)
class Guider_ImageGuidedCFG(comfy.samplers.CFGGuider):
def set_cfg(self, model, cfg1, image_cfg, latent_img):
def set_cfg(self, model, cfg1, image_cfg, latent_img, img_weighting):
self.cfg1 = cfg1
self.icfg = image_cfg
self.img = latent_img
self.img_weighting = img_weighting
self.model = model
def set_conds(self, positive, negative):
@@ -571,16 +572,23 @@ class Guider_ImageGuidedCFG(comfy.samplers.CFGGuider):
out = comfy.samplers.calc_cond_batch(self.inner_model, [negative_cond, positive_cond], x, timestep, model_options)
img = self.img["samples"].to(out[1].device)
norm_out1 = torch.linalg.norm(out[1]) # Get norm of positive cond
res = img - out[1] * (out[1] / norm_out1 * (img / norm_out1)).sum() # Project positive cond onto image
res *= torch.linalg.norm(out[1]) / torch.linalg.norm(res) # Normalize to cond
res = self.model.model.model_sampling.calculate_denoised(timestep, res, out[1])
weight = 1.0 # Flat by default
match self.img_weighting:
case "flat":
weight = 1.0
case "linear down":
weight = (self.model.model.model_sampling.timestep(timestep) / 999.0)[:, None, None, None].clone()
cfg = comfy.samplers.cfg_function(self.inner_model, out[1], out[0], self.cfg1, x, timestep, model_options=model_options, cond=positive_cond, uncond=negative_cond)
return cfg + (cfg - res) * self.icfg / self.cfg1 / 10 # Divide by 10 to mimic user-cfg. Do CFG - Res since the image is inverted the other way around.
return cfg + (cfg - res) * self.icfg / self.cfg1 / 10 * weight # Divide by 10 to mimic user-cfg. Do CFG - Res since the image is inverted the other way around.
class ImageGuidedCFGGuider:
@classmethod
@@ -591,6 +599,7 @@ class ImageGuidedCFGGuider:
"negative": ("CONDITIONING", ),
"cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0, "step":0.1, "round": 0.01}),
"image_cfg": ("FLOAT", {"default": 0.1, "min": -100.0, "max": 100.0, "step":0.1, "round": 0.01}),
"image_weighting": (["flat", "linear down"], ),
"latent_image": ("LATENT", ),
}
}
@@ -600,10 +609,10 @@ class ImageGuidedCFGGuider:
FUNCTION = "get_guider"
CATEGORY = "sampling/custom_sampling/guiders"
def get_guider(self, model, positive, negative, cfg, image_cfg, latent_image):
def get_guider(self, model, positive, negative, cfg, image_cfg, image_weighting, latent_image):
guider = Guider_ImageGuidedCFG(model)
guider.set_conds(positive, negative) # Conds
guider.set_cfg(model, cfg, image_cfg, latent_image) # Strengths
guider.set_cfg(model, cfg, image_cfg, latent_image, image_weighting) # Strengths
return (guider,)
class Guider_ScaledCFG(comfy.samplers.CFGGuider):