Update handler.py

handler.py

@@ -1,6 +1,6 @@
 from typing import  Dict, List, Any
 import torch
-from diffusers import DPMSolverMultistepScheduler, StableDiffusionInpaintPipeline, AutoPipelineForInpainting, AutoPipelineForImage2Image, StableDiffusionXLImg2ImgPipeline
+from diffusers import StableDiffusionXLImg2ImgPipeline
 from PIL import Image
 import base64
 from io import BytesIO
@@ -15,12 +15,8 @@ if device.type != 'cuda':
 class EndpointHandler():
     def __init__(self, path=""):
 
-        #self.fast_pipe = AutoPipelineForInpainting.from_pretrained("diffusers/stable-diffusion-xl-1.0-inpainting-0.1", torch_dtype=torch.float16, variant="fp16").to("cuda")
-        #self.generator = torch.Generator(device="cuda").manual_seed(0)
-
-
         self.smooth_pipe = StableDiffusionXLImg2ImgPipeline.from_pretrained(
-          "stabilityai/stable-diffusion-xl-refiner-1.0", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
+          "stabilityai/stable-diffusion-xl-refiner-1.0", torch_dtype=torch.float16
         )
         self.smooth_pipe.to("cuda")
         
@@ -38,7 +34,8 @@ class EndpointHandler():
 
         if encoded_image is not None:
             image = self.decode_base64_image(encoded_image)
-            out = self.smooth_pipe(prompt, image=image).images[0]
+            self.smooth_pipe.enable_xformers_memory_efficient_attention()
+            out = self.smooth_pipe(prompt, image=image, ).images[0]
 
             return out