Update app.py

app.py

@@ -1,41 +1,307 @@
+#!/usr/bin/env python
+
+import os
+import random
+import uuid
+import json
+
 import gradio as gr
+import numpy as np
+from PIL import Image
+import spaces
 import torch
-from diffusers import StableDiffusionPipeline
-import uuid
+from diffusers import DiffusionPipeline
+from typing import Tuple
+
+# Check for the Model Base..//
+
+bad_words = json.loads(os.getenv('BAD_WORDS', '["violence", "blood", "scary"]'))
+bad_words_negative = json.loads(os.getenv('BAD_WORDS_NEGATIVE', "[]"))
+default_negative = os.getenv("default_negative","")
+
+def check_text(prompt, negative=""):
+    for i in bad_words:
+        if i in prompt:
+            return True
+    for i in bad_words_negative:
+        if i in negative:
+            return True
+    return False
+
+# Updated to child-friendly styles
+
+style_list = [
+
+    {
+        "name": "Cartoon",
+        "prompt": "colorful cartoon {prompt}. vibrant, playful, friendly, suitable for children, highly detailed, bright colors",
+        "negative_prompt": "scary, dark, violent, ugly, realistic",
+    },
+    {
+        "name": "Children's Illustration",
+        "prompt": "children's illustration {prompt}. cute, colorful, fun, simple shapes, smooth lines, highly detailed, joyful",
+        "negative_prompt": "scary, dark, violent, deformed, ugly",
+    },   
+
+    {
+        "name": "Sticker",
+        "prompt": "children's sticker of {prompt}. bright colors, playful, high resolution, cartoonish",
+        "negative_prompt": "scary, dark, violent, ugly, low resolution",
+    },
+
+    {
+        "name": "Fantasy",
+        "prompt": "fantasy world for children with {prompt}. magical, vibrant, friendly, beautiful, colorful",
+        "negative_prompt": "dark, scary, violent, ugly, realistic",
+    },
+
+    {
+        "name": "(No style)",
+        "prompt": "{prompt}",
+        "negative_prompt": "",
+    },
+]
+
+styles = {k["name"]: (k["prompt"], k["negative_prompt"]) for k in style_list}
+STYLE_NAMES = list(styles.keys())
+DEFAULT_STYLE_NAME = "Sticker"
+
+def apply_style(style_name: str, positive: str, negative: str = "") -> Tuple[str, str]:
+    p, n = styles.get(style_name, styles[DEFAULT_STYLE_NAME])
+    if not negative:
+        negative = ""
+    return p.replace("{prompt}", positive), n + negative
+
+DESCRIPTION = """## Children's Sticker Generator
+
+Generate fun and playful stickers for children using AI.
+"""
+
+if not torch.cuda.is_available():
+    DESCRIPTION += "\n<p>⚠️Running on CPU, This may not work on CPU.</p>"
+
+MAX_SEED = np.iinfo(np.int32).max
+CACHE_EXAMPLES = torch.cuda.is_available() and os.getenv("CACHE_EXAMPLES", "0") == "1"
+MAX_IMAGE_SIZE = int(os.getenv("MAX_IMAGE_SIZE", "1024"))
+USE_TORCH_COMPILE = os.getenv("USE_TORCH_COMPILE", "0") == "1"
+ENABLE_CPU_OFFLOAD = os.getenv("ENABLE_CPU_OFFLOAD", "0") == "1"
 
-# Check for the Model Base
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
+NUM_IMAGES_PER_PROMPT = 1
+
 if torch.cuda.is_available():
-    pipe = StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4", torch_dtype=torch.float16, use_safetensors=True)
-    pipe.to(device)
-else:
-    pipe = StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4")
-
-def generate_stickers(word):
-    prompt = (f"Create a set of child-friendly stickers based on the word '{word}'. "
-              "Each sticker should have a black bold outline and fit within a label size of 70 x 35 mm, "
-              "and be organized into six distinct groups that fit a single A4 sheet with 24 labels per sheet. "
-              "Ensure the stickers do not overlap and the black outline is clearly visible. "
-              "The stickers should be fun, colorful, and cartoon-like, suitable for children, with a transparent or white background.")
-    negative_prompt = ("text, logos, watermarks, out of frame, ugly, extra limbs, bad anatomy, blurry, "
-                       "overlapping stickers, missing outline")
+    pipe = DiffusionPipeline.from_pretrained(
+        "SG161222/RealVisXL_V3.0_Turbo",
+        torch_dtype=torch.float16,
+        use_safetensors=True,
+        add_watermarker=False,
+        variant="fp16"
+    )
+    pipe2 = DiffusionPipeline.from_pretrained(
+        "SG161222/RealVisXL_V2.02_Turbo",
+        torch_dtype=torch.float16,
+        use_safetensors=True,
+        add_watermarker=False,
+        variant="fp16"
+    )
+    if ENABLE_CPU_OFFLOAD:
+        pipe.enable_model_cpu_offload()
+        pipe2.enable_model_cpu_offload()
+    else:
+        pipe.to(device)    
+        pipe2.to(device)    
+        print("Loaded on Device!")
+    
+    if USE_TORCH_COMPILE:
+        pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
+        pipe2.unet = torch.compile(pipe2.unet, mode="reduce-overhead", fullgraph=True)
+        print("Model Compiled!")
+
+def save_image(img):
+    unique_name = str(uuid.uuid4()) + ".png"
+    img.save(unique_name)
+    return unique_name
+
+def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
+    if randomize_seed:
+        seed = random.randint(0, MAX_SEED)
+    return seed
+
+@spaces.GPU(enable_queue=True)
+def generate(
+    prompt: str,
+    negative_prompt: str = "",
+    use_negative_prompt: bool = False,
+    style: str = DEFAULT_STYLE_NAME,
+    seed: int = 0,
+    width: int = 512,
+    height: int = 512,
+    guidance_scale: float = 3,
+    randomize_seed: bool = False,
+    use_resolution_binning: bool = True,
+    progress=gr.Progress(track_tqdm=True),
+):
+    if check_text(prompt, negative_prompt):
+        raise ValueError("Prompt contains restricted words.")
+    
+    prompt, negative_prompt = apply_style(style, prompt, negative_prompt)
+    seed = int(randomize_seed_fn(seed, randomize_seed))
+    generator = torch.Generator().manual_seed(seed)
+
+    if not use_negative_prompt:
+        negative_prompt = ""  # type: ignore
+    negative_prompt += default_negative    
+
+    options = {
+        "prompt": prompt,
+        "negative_prompt": negative_prompt,
+        "width": width,
+        "height": height,
+        "guidance_scale": guidance_scale,
+        "num_inference_steps": 25,
+        "generator": generator,
+        "num_images_per_prompt": NUM_IMAGES_PER_PROMPT,
+        "use_resolution_binning": use_resolution_binning,
+        "output_type": "pil",
+    }
     
-    width, height = 1024, 768  # Appropriate resolution for detailed stickers
-    images = pipe(prompt=prompt, negative_prompt=negative_prompt, width=width, height=height, num_inference_steps=50).images
-    output_paths = []
-    for image in images:
-        filename = f"{uuid.uuid4()}.png"
-        image.save(filename)
-        output_paths.append(filename)
-    return output_paths
-
-iface = gr.Interface(
-    fn=generate_stickers,
-    inputs="text",
-    outputs="image",
-    description="Enter a word to generate a set of child-friendly stickers that fit Sorex A4 Premium Sticker sheets (70x35mm labels per sheet)."
-)
+    images = pipe(**options).images + pipe2(**options).images
+
+    image_paths = [save_image(img) for img in images]
+    return image_paths, seed
+
+examples = [
+    "A cute cartoon bunny holding a carrot in a colorful garden",
+    "A playful dragon flying through the clouds, bright and friendly",
+    "A magical unicorn standing on a rainbow with sparkles",
+]
+
+css = '''
+.gradio-container{max-width: 700px !important}
+h1{text-align:center}
+'''
+
+with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
+    gr.Markdown(DESCRIPTION)
+    gr.DuplicateButton(
+        value="Duplicate Space for private use",
+        elem_id="duplicate-button",
+        visible=os.getenv("SHOW_DUPLICATE_BUTTON") == "1",
+    )
+    with gr.Group():
+        with gr.Row():
+            prompt = gr.Text(
+                label="Enter your prompt",
+                show_label=False,
+                max_lines=1,
+                placeholder="Enter a fun, child-friendly idea (e.g., cute bunny with a rainbow)",
+                container=False,
+            )
+            run_button = gr.Button("Run")
+        result = gr.Gallery(label="Generated Stickers", columns=1, preview=True)
+    with gr.Accordion("Advanced options", open=False):
+        use_negative_prompt = gr.Checkbox(label="Use negative prompt", value=True, visible=True)
+        negative_prompt = gr.Text(
+            label="Negative prompt",
+            max_lines=1,
+            placeholder="Enter a negative prompt",
+            value="(scary, violent, deformed, ugly, dark)",
+            visible=True,
+        )
+        with gr.Row():
+            num_inference_steps = gr.Slider(
+                label="Steps",
+                minimum=10,
+                maximum=60,
+                step=1,
+                value=25,
+            )
+        with gr.Row():
+            num_images_per_prompt = gr.Slider(
+                label="Images",
+                minimum=1,
+                maximum=5,
+                step=1,
+                value=2,
+            )
+        seed = gr.Slider(
+            label="Seed",
+            minimum=0,
+            maximum=MAX_SEED,
+            step=1,
+            value=0,
+            visible=True
+        )
+        randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+        with gr.Row(visible=True):
+            width = gr.Slider(
+                label="Width",
+                minimum=512,
+                maximum=1024,
+                step=8,
+                value=512,
+            )
+            height = gr.Slider(
+                label="Height",
+                minimum=512,
+                maximum=1024,
+                step=8,
+                value=512,
+            )
+        with gr.Row():
+            guidance_scale = gr.Slider(
+                label="Guidance Scale",
+                minimum=0.1,
+                maximum=20.0,
+                step=0.1,
+                value=7,
+            )
+    with gr.Row(visible=True):
+        style_selection = gr.Radio(
+            show_label=True,
+            container=True,
+            interactive=True,
+            choices=STYLE_NAMES,
+            value=DEFAULT_STYLE_NAME,
+            label="Sticker Style",
+        )
+    gr.Examples(
+        examples=examples,
+        inputs=prompt,
+        outputs=[result, seed],
+        fn=generate,
+        cache_examples=CACHE_EXAMPLES,
+    )
+
+    use_negative_prompt.change(
+        fn=lambda x: gr.update(visible=x),
+        inputs=use_negative_prompt,
+        outputs=negative_prompt,
+        api_name=False,
+    )
+
+    gr.on(
+        triggers=[
+            prompt.submit,
+            negative_prompt.submit,
+            run_button.click,
+        ],
+        fn=generate,
+        inputs=[
+            prompt,
+            negative_prompt,
+            use_negative_prompt,
+            style_selection,
+            seed,
+            width,
+            height,
+            guidance_scale,
+            randomize_seed,
+        ],
+        outputs=[result, seed],
+        api_name="run",
+    )
 
 if __name__ == "__main__":
-    iface.launch()
+    demo.queue(max_size=20).launch()