ComfyUI/comfy_extras/nodes_attention_multiply.py
Extraltodeus 8b41699da3
added optional layer filters
Comma separated, like this:
3,4,7,8

And if nothing is entered then it applies globally! :)
2024-05-16 00:55:06 +02:00

147 lines
7.2 KiB
Python

def attention_multiply(attn, model, q, k, v, out, block, layer):
m = model.clone()
sd = model.model_state_dict()
for key in sd:
if f"{block}_blocks.{layer}" not in key and block != "all":
continue
if key.endswith("{}.to_q.bias".format(attn)) or key.endswith("{}.to_q.weight".format(attn)):
m.add_patches({key: (None,)}, 0.0, q)
if key.endswith("{}.to_k.bias".format(attn)) or key.endswith("{}.to_k.weight".format(attn)):
m.add_patches({key: (None,)}, 0.0, k)
if key.endswith("{}.to_v.bias".format(attn)) or key.endswith("{}.to_v.weight".format(attn)):
m.add_patches({key: (None,)}, 0.0, v)
if key.endswith("{}.to_out.0.bias".format(attn)) or key.endswith("{}.to_out.0.weight".format(attn)):
m.add_patches({key: (None,)}, 0.0, out)
return m
class UNetSelfAttentionMultiply:
@classmethod
def INPUT_TYPES(s):
return {"required": { "model": ("MODEL",),
"q": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"k": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"v": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"out": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"block_id_input": ("STRING", {"multiline": False}, {"default": ""}),
"block_id_middle": ("STRING", {"multiline": False}, {"default": ""}),
"block_id_output": ("STRING", {"multiline": False}, {"default": ""}),
}}
RETURN_TYPES = ("MODEL",)
FUNCTION = "patch"
CATEGORY = "_for_testing/attention_experiments"
def patch(self, model, q, k, v, out, block_id_input, block_id_middle, block_id_output):
block_layers = {"input": block_id_input, "middle": block_id_middle, "output": block_id_output}
m = model.clone()
if all(value == "" for value in block_layers.values()):
m = attention_multiply("attn1", m, q, k, v, out, "all", "all")
else:
for block in block_layers:
for block_id in block_layers[block].split(","):
if block_id != "":
block_id = int(block_id)
m = attention_multiply("attn1", m, q, k, v, out, block, block_id)
return (m, )
class UNetCrossAttentionMultiply:
@classmethod
def INPUT_TYPES(s):
return {"required": { "model": ("MODEL",),
"q": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"k": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"v": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"out": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"block_id_input": ("STRING", {"multiline": False}, {"default": ""}),
"block_id_middle": ("STRING", {"multiline": False}, {"default": ""}),
"block_id_output": ("STRING", {"multiline": False}, {"default": ""}),
}}
RETURN_TYPES = ("MODEL",)
FUNCTION = "patch"
CATEGORY = "_for_testing/attention_experiments"
def patch(self, model, q, k, v, out, block_id_input, block_id_middle, block_id_output):
block_layers = {"input": block_id_input, "middle": block_id_middle, "output": block_id_output}
m = model.clone()
if all(value == "" for value in block_layers.values()):
m = attention_multiply("attn2", m, q, k, v, out, "all", "all")
else:
for block in block_layers:
for block_id in block_layers[block].split(","):
if block_id != "":
block_id = int(block_id)
m = attention_multiply("attn2", m, q, k, v, out, block, block_id)
return (m, )
class CLIPAttentionMultiply:
@classmethod
def INPUT_TYPES(s):
return {"required": { "clip": ("CLIP",),
"q": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"k": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"v": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"out": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
}}
RETURN_TYPES = ("CLIP",)
FUNCTION = "patch"
CATEGORY = "_for_testing/attention_experiments"
def patch(self, clip, q, k, v, out):
m = clip.clone()
sd = m.patcher.model_state_dict()
for key in sd:
if key.endswith("self_attn.q_proj.weight") or key.endswith("self_attn.q_proj.bias"):
m.add_patches({key: (None,)}, 0.0, q)
if key.endswith("self_attn.k_proj.weight") or key.endswith("self_attn.k_proj.bias"):
m.add_patches({key: (None,)}, 0.0, k)
if key.endswith("self_attn.v_proj.weight") or key.endswith("self_attn.v_proj.bias"):
m.add_patches({key: (None,)}, 0.0, v)
if key.endswith("self_attn.out_proj.weight") or key.endswith("self_attn.out_proj.bias"):
m.add_patches({key: (None,)}, 0.0, out)
return (m, )
class UNetTemporalAttentionMultiply:
@classmethod
def INPUT_TYPES(s):
return {"required": { "model": ("MODEL",),
"self_structural": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"self_temporal": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"cross_structural": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
"cross_temporal": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
}}
RETURN_TYPES = ("MODEL",)
FUNCTION = "patch"
CATEGORY = "_for_testing/attention_experiments"
def patch(self, model, self_structural, self_temporal, cross_structural, cross_temporal):
m = model.clone()
sd = model.model_state_dict()
for k in sd:
if (k.endswith("attn1.to_out.0.bias") or k.endswith("attn1.to_out.0.weight")):
if '.time_stack.' in k:
m.add_patches({k: (None,)}, 0.0, self_temporal)
else:
m.add_patches({k: (None,)}, 0.0, self_structural)
elif (k.endswith("attn2.to_out.0.bias") or k.endswith("attn2.to_out.0.weight")):
if '.time_stack.' in k:
m.add_patches({k: (None,)}, 0.0, cross_temporal)
else:
m.add_patches({k: (None,)}, 0.0, cross_structural)
return (m, )
NODE_CLASS_MAPPINGS = {
"UNetSelfAttentionMultiply": UNetSelfAttentionMultiply,
"UNetCrossAttentionMultiply": UNetCrossAttentionMultiply,
"CLIPAttentionMultiply": CLIPAttentionMultiply,
"UNetTemporalAttentionMultiply": UNetTemporalAttentionMultiply,
}