Update app.py

app.py

@@ -1,29 +1,53 @@
 import gradio as gr
 import numpy as np
 import random
-
-# import spaces #[uncomment to use ZeroGPU]
-from diffusers import DiffusionPipeline
 import torch
+from diffusers import DiffusionPipeline
 
-device = "cuda" if torch.cuda.is_available() else "cpu"
-model_repo_id = "stabilityai/sdxl-turbo"  # Replace to the model you would like to use
+# === Configuration ===
+MODEL_REPO_ID = "stabilityai/sdxl-turbo"
+MAX_SEED = np.iinfo(np.int32).max
+MAX_IMAGE_SIZE = 1024
 
-if torch.cuda.is_available():
-    torch_dtype = torch.float16
-else:
-    torch_dtype = torch.float32
+def get_torch_dtype():
+    return torch.float16 if torch.cuda.is_available() else torch.float32
 
-pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
+def get_device():
+    return "cuda" if torch.cuda.is_available() else "cpu"
+
+# === Load the diffusion model ===
+torch_dtype = get_torch_dtype()
+device = get_device()
+pipe = DiffusionPipeline.from_pretrained(MODEL_REPO_ID, torch_dtype=torch_dtype)
 pipe = pipe.to(device)
 
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
+# === Define custom prompt builder ===
+def build_prompt(word):
+    return (
+        f"Create a powerful, emotionally resonant image that vividly illustrates the meaning of the word '{word}', "
+        f"so that even someone who doesn’t speak English can understand it instantly. "
+        f"The visual should be sharp, symbolic, and universally relatable. "
+        f"Seamlessly weave the word '{word}' into the scene—clearly spelled but not overpowering—"
+        f"so it supports the concept without drawing attention away. "
+        f"Format: 1080x1080 pixels (square) for Instagram in a (.png) format."
+    )
 
+# === Image generation function ===
+def generate_image(prompt, negative_prompt, guidance_scale, num_inference_steps, width, height, seed):
+    generator = torch.Generator().manual_seed(seed)
+    return 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]
 
-# @spaces.GPU #[uncomment to use ZeroGPU]
+# === Inference wrapper ===
 def infer(
-    prompt,
+    word,
     negative_prompt,
     seed,
     randomize_seed,
@@ -36,27 +60,10 @@ def infer(
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
 
-    generator = torch.Generator().manual_seed(seed)
-
-    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]
-
+    prompt = build_prompt(word)
+    image = generate_image(prompt, negative_prompt, guidance_scale, num_inference_steps, width, height, seed)
     return image, seed
 
-
-examples = [
-    "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
-    "An astronaut riding a green horse",
-    "A delicious ceviche cheesecake slice",
-]
-
 css = """
 #col-container {
     margin: 0 auto;
@@ -66,20 +73,19 @@ css = """
 
 with gr.Blocks(css=css) as demo:
     with gr.Column(elem_id="col-container"):
-        gr.Markdown(" # Text-to-Image Gradio Template")
+        gr.Markdown(" # Word-to-Image Generator for Instagram 🎨")
 
         with gr.Row():
-            prompt = gr.Text(
-                label="Prompt",
+            word = gr.Text(
+                label="Vocabulary Word",
                 show_label=False,
                 max_lines=1,
-                placeholder="Enter your prompt",
+                placeholder="Enter a vocabulary word",
                 container=False,
             )
+            run_button = gr.Button("Generate Image", scale=0, variant="primary")
 
-            run_button = gr.Button("Run", scale=0, variant="primary")
-
-        result = gr.Image(label="Result", show_label=False)
+        result = gr.Image(label="Generated Image", show_label=False)
 
         with gr.Accordion("Advanced Settings", open=False):
             negative_prompt = gr.Text(
@@ -88,65 +94,20 @@ with gr.Blocks(css=css) as demo:
                 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=MAX_SEED, step=1, value=0)
             randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
 
             with gr.Row():
-                width = gr.Slider(
-                    label="Width",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=1024,  # Replace with defaults that work for your model
-                )
-
-                height = gr.Slider(
-                    label="Height",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=1024,  # Replace with defaults that work for your model
-                )
+                width = gr.Slider(label="Width", minimum=256, maximum=MAX_IMAGE_SIZE, step=32, value=1080)
+                height = gr.Slider(label="Height", minimum=256, maximum=MAX_IMAGE_SIZE, step=32, value=1080)
 
             with gr.Row():
-                guidance_scale = gr.Slider(
-                    label="Guidance scale",
-                    minimum=0.0,
-                    maximum=10.0,
-                    step=0.1,
-                    value=0.0,  # Replace with defaults that work for your model
-                )
-
-                num_inference_steps = gr.Slider(
-                    label="Number of inference steps",
-                    minimum=1,
-                    maximum=50,
-                    step=1,
-                    value=2,  # Replace with defaults that work for your model
-                )
-
-        gr.Examples(examples=examples, inputs=[prompt])
-    gr.on(
-        triggers=[run_button.click, prompt.submit],
+                guidance_scale = gr.Slider(label="Guidance scale", minimum=0.0, maximum=10.0, step=0.1, value=3.5)
+                num_inference_steps = gr.Slider(label="Inference steps", minimum=1, maximum=50, step=1, value=20)
+
+    run_button.click(
         fn=infer,
-        inputs=[
-            prompt,
-            negative_prompt,
-            seed,
-            randomize_seed,
-            width,
-            height,
-            guidance_scale,
-            num_inference_steps,
-        ],
+        inputs=[word, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
         outputs=[result, seed],
     )