Update app.py

app.py

@@ -3,12 +3,11 @@ import numpy as np
 import random
 from diffusers import DiffusionPipeline
 import torch
-from torch.fft import fftn, ifftn  # For Fourier transforms
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
-# Load the model
 if torch.cuda.is_available():
+    torch.cuda.max_memory_allocated(device=device)
     pipe = DiffusionPipeline.from_pretrained("stabilityai/sdxl-turbo", torch_dtype=torch.float16, variant="fp16", use_safetensors=True)
     pipe.enable_xformers_memory_efficient_attention()
     pipe = pipe.to(device)
@@ -19,42 +18,14 @@ else:
 MAX_SEED = np.iinfo(np.int32).max
 MAX_IMAGE_SIZE = 1024
 
-# Group-theory-based seed reduction (example: cyclic group on mod N)
-def reduce_seeds(seed):
-    group_size = 100  # Cyclic group of size 100
-    return seed % group_size
-
-# Fourier-based optimization with proper tensor conversion
-def fft_convolution(image):
-    # Convert the image to a PyTorch tensor
-    tensor_image = torch.tensor(image).permute(2, 0, 1).float()  # Convert to tensor and adjust channels
-
-    # Convert to frequency domain using FFT
-    freq_image = fftn(tensor_image)
-
-    # Apply some transformation in the frequency domain (example: filtering, smoothing)
-    # This is a placeholder; implement any frequency domain operation here
-    transformed_freq_image = freq_image * torch.exp(-torch.abs(freq_image))  # Example operation
-
-    # Convert back to spatial domain using inverse FFT
-    transformed_image = ifftn(transformed_freq_image).real
-
-    # Convert back to NumPy array and adjust channels back to original shape
-    result_image = transformed_image.permute(1, 2, 0).numpy()
-    
-    return result_image
-
 def infer(prompt_part1, color, dress_type, design, prompt_part5, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps):
     prompt = f"{prompt_part1} {color} colored plain {dress_type} with {design} design, {prompt_part5}"
     
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
-    else:
-        seed = reduce_seeds(seed)  # Apply symmetry-based seed reduction
         
     generator = torch.Generator().manual_seed(seed)
     
-    # Run the diffusion model
     image = pipe(
         prompt=prompt, 
         negative_prompt=negative_prompt,
@@ -63,13 +34,9 @@ def infer(prompt_part1, color, dress_type, design, prompt_part5, negative_prompt
         width=width, 
         height=height,
         generator=generator
-    ).images[0]
-
-    # Apply Fourier-based convolution optimization
-    image = np.array(image)  # Convert to numpy array
-    optimized_image = fft_convolution(image)  # Apply Fourier convolution
-
-    return optimized_image
+    ).images[0] 
+    
+    return image
 
 examples = [
     "red, t-shirt, yellow stripes",
@@ -143,6 +110,7 @@ with gr.Blocks(css=css) as demo:
                 visible=False,
             )
             
+            
             run_button = gr.Button("Run", scale=0)
         
         result = gr.Image(label="Result", show_label=False)
@@ -213,4 +181,4 @@ with gr.Blocks(css=css) as demo:
         outputs=[result]
     )
 
-demo.queue().launch()
+demo.queue().launch()