Merge pull request #338 from xiaohu2015/main

for the new diffuser version

ip_adapter/attention_processor.py

@@ -23,6 +23,8 @@ class AttnProcessor(nn.Module):
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 
@@ -109,6 +111,8 @@ class IPAttnProcessor(nn.Module):
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 
@@ -205,6 +209,8 @@ class AttnProcessor2_0(torch.nn.Module):
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 
@@ -308,6 +314,8 @@ class IPAttnProcessor2_0(torch.nn.Module):
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 
@@ -413,7 +421,7 @@ class CNAttnProcessor:
     def __init__(self, num_tokens=4):
         self.num_tokens = num_tokens
 
-    def __call__(self, attn, hidden_states, encoder_hidden_states=None, attention_mask=None, temb=None):
+    def __call__(self, attn, hidden_states, encoder_hidden_states=None, attention_mask=None, temb=None, *args, **kwargs,):
         residual = hidden_states
 
         if attn.spatial_norm is not None:
@@ -487,6 +495,8 @@ class CNAttnProcessor2_0:
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 

ip_adapter/attention_processor_faceid.py

@@ -36,6 +36,8 @@ class LoRAAttnProcessor(nn.Module):
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 
@@ -130,6 +132,8 @@ class LoRAIPAttnProcessor(nn.Module):
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 
@@ -236,6 +240,8 @@ class LoRAAttnProcessor2_0(nn.Module):
         encoder_hidden_states=None,
         attention_mask=None,
         temb=None,
+        *args,
+        **kwargs,
     ):
         residual = hidden_states
 
@@ -335,7 +341,7 @@ class LoRAIPAttnProcessor2_0(nn.Module):
         self.to_v_ip = nn.Linear(cross_attention_dim or hidden_size, hidden_size, bias=False)
 
     def __call__(
-        self, attn, hidden_states, encoder_hidden_states=None, attention_mask=None, scale=1.0, temb=None
+        self, attn, hidden_states, encoder_hidden_states=None, attention_mask=None, scale=1.0, temb=None, *args, **kwargs,
     ):
         residual = hidden_states