Update app.py

app.py

@@ -1,46 +1,61 @@
 import gradio as gr
-import numpy as np
-import random
-from diffusers import DiffusionPipeline
-import torch
+import replicate
+from openai import OpenAI
+from PIL import Image
+import requests
+from io import BytesIO
 
-device = "cuda" if torch.cuda.is_available() else "cpu"
 
-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)
-else: 
-    pipe = DiffusionPipeline.from_pretrained("stabilityai/sdxl-turbo", use_safetensors=True)
-    pipe = pipe.to(device)
+def generate_image_openai(prompt):
 
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
+    client = OpenAI()
+    response = client.images.generate(
+        model="dall-e-3",
+        prompt=prompt,
+        size="1024x1024",
+        n=1,
+    )
 
-def infer(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps):
+    return response.data[0].url, response.data[0].revised_prompt
 
-    if randomize_seed:
-        seed = random.randint(0, MAX_SEED)
-        
-    generator = torch.Generator().manual_seed(seed)
+def style_transfer(input_image_path, style_image_path, revised_prompt):
+    input = {
+        # "image": open(input_image_path, "rb"),
+        "image": input_image_path,
+        "image_style": open(style_image_path, "rb"),
+        "style_strength": 0.4,
+        "structure_strength":1.2,
+        "prompt": " natural light, natural bright colors, low quality, candid, grainy, instagram photo, phone camera, candid, blurry home video, high iso noisy" ,
+        "seed": 42,
+    }
+    output = replicate.run(
+        "prakharsaxena24/2d-to-real-style:c0e1e612a11a13d1d57a6d647e7665ad850bc73715337c1f499bb7b52404c35a",
+        input=input
+    )
+    return output[0]
+
+
+
+def infer(text,title):
+
+    prompt = f"""Please create a simple suitable image to accompany the following text as part of an article with the title "{title}". The objects in the image must have realistic proportions. Text: "{text}" 
+     Please make sure not to include text in the image."""
+    image_url_openai, revised_prompt = generate_image_openai(prompt)
+    style_image_url = style_transfer(image_url_openai, f'./style.png',revised_prompt)
+    response = requests.get(style_image_url)
+    img = Image.open(BytesIO(response.content))
+
+    return img
+    
+    
     
-    image = pipe(
-        prompt = prompt, 
-        negative_prompt = negative_prompt,
-        guidance_scale = guidance_scale, 
-        num_inference_steps = num_inference_steps, 
-        width = width, 
-        height = height,
-        generator = generator
-    ).images[0] 
     
-    return image
+
 
 examples = [
-    "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
-    "An astronaut riding a green horse",
-    "A delicious ceviche cheesecake slice",
+    # "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
+    # "An astronaut riding a green horse",
+    # "A delicious ceviche cheesecake slice",
 ]
 
 css="""
@@ -50,10 +65,7 @@ css="""
 }
 """
 
-if torch.cuda.is_available():
-    power_device = "GPU"
-else:
-    power_device = "CPU"
+
 
 with gr.Blocks(css=css) as demo:
     
@@ -65,8 +77,15 @@ with gr.Blocks(css=css) as demo:
         
         with gr.Row():
             
-            prompt = gr.Text(
-                label="Prompt",
+            text = gr.Text(
+                label="Text",
+                show_label=False,
+                max_lines=1,
+                placeholder="Enter your prompt",
+                container=False,
+            )
+            title = gr.Text(
+                label="Title",
                 show_label=False,
                 max_lines=1,
                 placeholder="Enter your prompt",
@@ -77,60 +96,60 @@ with gr.Blocks(css=css) as demo:
         
         result = gr.Image(label="Result", show_label=False)
 
-        with gr.Accordion("Advanced Settings", open=False):
+        # with gr.Accordion("Advanced Settings", open=False):
             
-            negative_prompt = gr.Text(
-                label="Negative prompt",
-                max_lines=1,
-                placeholder="Enter a negative prompt",
-                visible=False,
-            )
+        #     negative_prompt = gr.Text(
+        #         label="Negative prompt",
+        #         max_lines=1,
+        #         placeholder="Enter a negative prompt",
+        #         visible=False,
+        #     )
             
-            seed = gr.Slider(
-                label="Seed",
-                minimum=0,
-                maximum=MAX_SEED,
-                step=1,
-                value=0,
-            )
+        #     seed = gr.Slider(
+        #         label="Seed",
+        #         minimum=0,
+        #         maximum=100000,
+        #         step=1,
+        #         value=0,
+        #     )
             
-            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+        #     randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
             
-            with gr.Row():
+        #     with gr.Row():
                 
-                width = gr.Slider(
-                    label="Width",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=512,
-                )
+        #         width = gr.Slider(
+        #             label="Width",
+        #             minimum=256,
+        #             maximum=MAX_IMAGE_SIZE,
+        #             step=32,
+        #             value=512,
+        #         )
                 
-                height = gr.Slider(
-                    label="Height",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=512,
-                )
+        #         height = gr.Slider(
+        #             label="Height",
+        #             minimum=256,
+        #             maximum=MAX_IMAGE_SIZE,
+        #             step=32,
+        #             value=512,
+        #         )
             
-            with gr.Row():
+        #     with gr.Row():
                 
-                guidance_scale = gr.Slider(
-                    label="Guidance scale",
-                    minimum=0.0,
-                    maximum=10.0,
-                    step=0.1,
-                    value=0.0,
-                )
+        #         guidance_scale = gr.Slider(
+        #             label="Guidance scale",
+        #             minimum=0.0,
+        #             maximum=10.0,
+        #             step=0.1,
+        #             value=0.0,
+        #         )
                 
-                num_inference_steps = gr.Slider(
-                    label="Number of inference steps",
-                    minimum=1,
-                    maximum=12,
-                    step=1,
-                    value=2,
-                )
+        #         num_inference_steps = gr.Slider(
+        #             label="Number of inference steps",
+        #             minimum=1,
+        #             maximum=12,
+        #             step=1,
+        #             value=2,
+        #         )
         
         gr.Examples(
             examples = examples,
@@ -139,7 +158,7 @@ with gr.Blocks(css=css) as demo:
 
     run_button.click(
         fn = infer,
-        inputs = [prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
+        inputs = [text, title],
         outputs = [result]
     )