Update app.py

app.py

@@ -11,19 +11,17 @@ torch.cuda.empty_cache()
 
 def genie (Model, Prompt, negative_prompt, height, width, scale, steps, seed, upscale, high_noise_frac):
     generator = np.random.seed(0) if seed == 0 else torch.manual_seed(seed)
-    
-    if upscale == "Yes":
-        refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
-        refiner.enable_xformers_memory_efficient_attention()
-        refiner = refiner.to(device)
-        torch.cuda.empty_cache()
-        
+       
     if Model == "PhotoReal":
         pipe = DiffusionPipeline.from_pretrained("circulus/canvers-real-v3.8.1", torch_dtype=torch.float16, safety_checker=None) if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("circulus/canvers-real-v3.8.1")
         pipe.enable_xformers_memory_efficient_attention()
         pipe = pipe.to(device)
         torch.cuda.empty_cache()
         if upscale == "Yes":
+            refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
+            refiner.enable_xformers_memory_efficient_attention()
+            refiner = refiner.to(device)
+            torch.cuda.empty_cache()
             int_image = pipe(Prompt, negative_prompt=negative_prompt, height=height, width=width, num_inference_steps=steps, guidance_scale=scale).images
             image = refiner(Prompt, negative_prompt=negative_prompt, image=int_image, denoising_start=high_noise_frac).images[0]
             torch.cuda.empty_cache()
@@ -39,6 +37,10 @@ def genie (Model, Prompt, negative_prompt, height, width, scale, steps, seed, up
         anime = anime.to(device)
         torch.cuda.empty_cache()
         if upscale == "Yes":
+            refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
+            refiner.enable_xformers_memory_efficient_attention()
+            refiner = refiner.to(device)
+            torch.cuda.empty_cache()
             int_image = anime(Prompt, negative_prompt=negative_prompt, height=height, width=width, num_inference_steps=steps, guidance_scale=scale).images
             image = refiner(Prompt, negative_prompt=negative_prompt, image=int_image, denoising_start=high_noise_frac).images[0]
             torch.cuda.empty_cache()
@@ -54,6 +56,10 @@ def genie (Model, Prompt, negative_prompt, height, width, scale, steps, seed, up
         disney = disney.to(device)
         torch.cuda.empty_cache()
         if upscale == "Yes":
+            refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
+            refiner.enable_xformers_memory_efficient_attention()
+            refiner = refiner.to(device)
+            torch.cuda.empty_cache()
             int_image = disney(Prompt, negative_prompt=negative_prompt, height=height, width=width, num_inference_steps=steps, guidance_scale=scale).images
             image = refiner(Prompt, negative_prompt=negative_prompt, image=int_image, denoising_start=high_noise_frac).images[0]
             torch.cuda.empty_cache()
@@ -69,6 +75,10 @@ def genie (Model, Prompt, negative_prompt, height, width, scale, steps, seed, up
         story = story.to(device)
         torch.cuda.empty_cache()
         if upscale == "Yes":
+            refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
+            refiner.enable_xformers_memory_efficient_attention()
+            refiner = refiner.to(device)
+            torch.cuda.empty_cache()
             int_image = story(Prompt, negative_prompt=negative_prompt, height=height, width=width, num_inference_steps=steps, guidance_scale=scale).images
             image = refiner(Prompt, negative_prompt=negative_prompt, image=int_image, denoising_start=high_noise_frac).images[0]
             torch.cuda.empty_cache()
@@ -84,6 +94,10 @@ def genie (Model, Prompt, negative_prompt, height, width, scale, steps, seed, up
         semi = semi.to(device)
         torch.cuda.empty_cache()
         if upscale == "Yes":
+            refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
+            refiner.enable_xformers_memory_efficient_attention()
+            refiner = refiner.to(device)
+            torch.cuda.empty_cache()
             image = semi(Prompt, negative_prompt=negative_prompt, height=height, width=width, num_inference_steps=steps, guidance_scale=scale, output_type="latent").images
             image = refiner(Prompt, negative_prompt=negative_prompt, image=image, denoising_start=high_noise_frac).images[0]
             torch.cuda.empty_cache()
@@ -99,6 +113,10 @@ def genie (Model, Prompt, negative_prompt, height, width, scale, steps, seed, up
         animagine = animagine.to(device)
         torch.cuda.empty_cache()
         if upscale == "Yes":
+            refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
+            refiner.enable_xformers_memory_efficient_attention()
+            refiner = refiner.to(device)
+            torch.cuda.empty_cache()
             int_image = animagine(Prompt, negative_prompt=negative_prompt, height=height, width=width, num_inference_steps=steps, guidance_scale=scale, output_type="latent").images
             image = refiner(Prompt, negative_prompt=negative_prompt, image=int_image, denoising_start=high_noise_frac).images[0]
             torch.cuda.empty_cache()
@@ -117,8 +135,12 @@ def genie (Model, Prompt, negative_prompt, height, width, scale, steps, seed, up
         torch.cuda.empty_cache()
     
         if upscale == "Yes":
-            torch.cuda.empty_cache()
             torch.cuda.max_memory_allocated(device=device)
+            torch.cuda.empty_cache()
+            refiner = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0", use_safetensors=True, torch_dtype=torch.float16, variant="fp16") if torch.cuda.is_available() else DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-1.0")
+            refiner.enable_xformers_memory_efficient_attention()
+            refiner = refiner.to(device)
+            torch.cuda.empty_cache()
             image = sdxl(Prompt, negative_prompt=negative_prompt, height=height, width=width, num_inference_steps=steps, guidance_scale=scale, output_type="latent").images
             torch.cuda.empty_cache()
             refined = refiner(Prompt, negative_prompt=negative_prompt, image=image, denoising_start=high_noise_frac).images[0]