Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

Mask strength should be separate from area strength.

comfyanonymous071011aebed2
diff --git a/comfy/samplers.py b/comfy/samplers.py
index f8701c879..10527fb1c 100644
--- a/comfy/samplers.py
+++ b/comfy/samplers.py
@@ -26,10 +26,13 @@ def sampling_function(model_function, x, timestep, uncond, cond, cond_scale, con
if 'mask' in cond[1]:
# Scale the mask to the size of the input
# The mask should have been resized as we began the sampling process
+ mask_strength = 1.0
+ if "mask_strength" in cond[1]:
+ mask_strength = cond[1]["mask_strength"]
mask = cond[1]['mask']
assert(mask.shape[1] == x_in.shape[2])
assert(mask.shape[2] == x_in.shape[3])
- mask = mask[:,area[2]:area[0] + area[2],area[3]:area[1] + area[3]]
+ mask = mask[:,area[2]:area[0] + area[2],area[3]:area[1] + area[3]] * mask_strength
mask = mask.unsqueeze(1).repeat(input_x.shape[0] // mask.shape[0], input_x.shape[1], 1, 1)
else:
mask = torch.ones_like(input_x)
diff --git a/nodes.py b/nodes.py
index 12fa7e5a3..b4069c836 100644
--- a/nodes.py
+++ b/nodes.py
@@ -98,7 +98,7 @@ class ConditioningSetMask:
CATEGORY = "conditioning"
- def append(self, conditioning, mask, set_area_to_bounds, strength, min_sigma=0.0, max_sigma=99.0):
+ def append(self, conditioning, mask, set_area_to_bounds, strength):
c = []
if len(mask.shape) < 3:
mask = mask.unsqueeze(0)
@@ -107,9 +107,7 @@ class ConditioningSetMask:
_, h, w = mask.shape
n[1]['mask'] = mask
n[1]['set_area_to_bounds'] = set_area_to_bounds
- n[1]['strength'] = strength
- n[1]['min_sigma'] = min_sigma
- n[1]['max_sigma'] = max_sigma
+ n[1]['mask_strength'] = strength
c.append(n)
return (c, )