s

handler.py

@@ -20,16 +20,41 @@ dtype = torch.bfloat16 if torch.cuda.get_device_capability()[0] == 8 else torch.
 
 class EndpointHandler():
      def __init__(self, path=""):
-         self.stable_diffusion_id = "Lykon/dreamshaper-8"
+        self.stable_diffusion_id = "Lykon/dreamshaper-8"
 
 
 
 
 
-         self.pipe = StableDiffusionPipeline.from_pretrained(self.stable_diffusion_id,torch_dtype=dtype,safety_checker=StableDiffusionSafetyChecker.from_pretrained("CompVis/stable-diffusion-safety-checker", torch_dtype=dtype)).to(device.type)
-         #self.pipe.enable_xformers_memory_efficient_attention()
-         #self.pipe.enable_vae_tiling()
-         self.generator = torch.Generator(device=device.type).manual_seed(3)
+        self.pipe = StableDiffusionPipeline.from_pretrained(self.stable_diffusion_id,torch_dtype=dtype,safety_checker=StableDiffusionSafetyChecker.from_pretrained("CompVis/stable-diffusion-safety-checker", torch_dtype=dtype)).to(device.type)
+        #self.pipe.enable_xformers_memory_efficient_attention()
+        #self.pipe.enable_vae_tiling()
+        self.generator = torch.Generator(device=device.type).manual_seed(3)
+
+
+        from typing import Optional
+        from torch import Tensor
+        from torch.nn import functional as F
+        from torch.nn import Conv2d
+        from torch.nn.modules.utils import _pair
+
+        def asymmetricConv2DConvForward(self, input: Tensor, weight: Tensor, bias: Optional[Tensor]):
+            self.paddingX = (self._reversed_padding_repeated_twice[0], self._reversed_padding_repeated_twice[1], 0, 0)
+            self.paddingY = (0, 0, self._reversed_padding_repeated_twice[2], self._reversed_padding_repeated_twice[3])
+            working = F.pad(input, self.paddingX, mode='circular')
+            working = F.pad(working, self.paddingY, mode='constant')
+            return F.conv2d(working, weight, bias, self.stride, _pair(0), self.dilation, self.groups)
+
+        targets = [pipe.vae, pipe.text_encoder, pipe.unet,]
+        conv_layers = []
+        for target in targets:
+            for module in target.modules():
+                if isinstance(module, torch.nn.Conv2d):
+                    conv_layers.append(module)
+
+        for cl in conv_layers:
+            cl._conv_forward = asymmetricConv2DConvForward.__get__(cl, Conv2d)
+
 
      def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
 #         """