Update app.py

app.py

@@ -1,174 +1,44 @@
-from __future__ import annotations
-import os
-import random
-import uuid
-import gradio as gr
 import spaces
-import numpy as np
-import uuid
-from diffusers import PixArtAlphaPipeline, LCMScheduler
-import torch
-from typing import Tuple
-from datetime import datetime
-
-
-DESCRIPTION = """ # Instant Image
-        ### Super fast text to Image Generator.
-        ### <span style='color: red;'>You may change the steps from 4 to 8, if you didn't get satisfied results.
-        ### First Image processing takes time then images generate faster.
-        """
-if not torch.cuda.is_available():
-    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
-
-MAX_SEED = np.iinfo(np.int32).max
-CACHE_EXAMPLES = torch.cuda.is_available() and os.getenv("CACHE_EXAMPLES", "1") == "1"
-MAX_IMAGE_SIZE = int(os.getenv("MAX_IMAGE_SIZE", "3000"))
-USE_TORCH_COMPILE = os.getenv("USE_TORCH_COMPILE", "0") == "1"
-ENABLE_CPU_OFFLOAD = os.getenv("ENABLE_CPU_OFFLOAD", "0") == "1"
-PORT = int(os.getenv("DEMO_PORT", "15432"))
-
-device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
-
-
-style_list = [
-    {
-        "name": "(No style)",
-        "prompt": "{prompt}",
-        "negative_prompt": "",
-    },
-    {
-        "name": "Cinematic",
-        "prompt": "cinematic still {prompt} . emotional, harmonious, vignette, highly detailed, high budget, bokeh, cinemascope, moody, epic, gorgeous, film grain, grainy",
-        "negative_prompt": "anime, cartoon, graphic, text, painting, crayon, graphite, abstract, glitch, deformed, mutated, ugly, disfigured",
-    },
-    {
-        "name": "Realistic",
-        "prompt": "Photorealistic {prompt} . Ulta-realistic, professional, 4k, highly detailed",
-        "negative_prompt": "drawing, painting, crayon, sketch, graphite, impressionist, noisy, blurry, soft, deformed, ugly, disfigured",
-    },
-    {
-        "name": "Anime",
-        "prompt": "anime artwork {prompt} . anime style, key visual, vibrant, studio anime,  highly detailed",
-        "negative_prompt": "photo, deformed, black and white, realism, disfigured, low contrast",
-    },
-    {
-        "name": "Digital Art",
-        "prompt": "concept art {prompt} . digital artwork, illustrative, painterly, matte painting, highly detailed",
-        "negative_prompt": "photo, photorealistic, realism, ugly",
-    },
-    {
-        "name": "Pixel art",
-        "prompt": "pixel-art {prompt} . low-res, blocky, pixel art style, 8-bit graphics",
-        "negative_prompt": "sloppy, messy, blurry, noisy, highly detailed, ultra textured, photo, realistic",
-    },
-    {
-        "name": "Fantasy art",
-        "prompt": "ethereal fantasy concept art of  {prompt} . magnificent, celestial, ethereal, painterly, epic, majestic, magical, fantasy art, cover art, dreamy",
-        "negative_prompt": "photographic, realistic, realism, 35mm film, dslr, cropped, frame, text, deformed, glitch, noise, noisy, off-center, deformed, cross-eyed, closed eyes, bad anatomy, ugly, disfigured, sloppy, duplicate, mutated, black and white",
-    },
-    {
-        "name": "3D Model",
-        "prompt": "professional 3d model {prompt} . octane render, highly detailed, volumetric, dramatic lighting",
-        "negative_prompt": "ugly, deformed, noisy, low poly, blurry, painting",
-    },
-]
-
-
-styles = {k["name"]: (k["prompt"], k["negative_prompt"]) for k in style_list}
-STYLE_NAMES = list(styles.keys())
-DEFAULT_STYLE_NAME = "(No style)"
-NUM_IMAGES_PER_PROMPT = 1
-
-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
-
-if torch.cuda.is_available():
-
-    pipe = PixArtAlphaPipeline.from_pretrained(
-        "PixArt-alpha/PixArt-LCM-XL-2-1024-MS",
-        torch_dtype=torch.float16,
-        use_safetensors=True,
-    )
-
-    if os.getenv('CONSISTENCY_DECODER', False):
-        print("Using DALL-E 3 Consistency Decoder")
-        pipe.vae = ConsistencyDecoderVAE.from_pretrained("openai/consistency-decoder", torch_dtype=torch.float16)
-
-    if ENABLE_CPU_OFFLOAD:
-        pipe.enable_model_cpu_offload()
-    else:
-        pipe.to(device)
-        print("Loaded on Device!")
-        
-    # speed-up T5
-    pipe.text_encoder.to_bettertransformer()
+import argparse
+import os
+import time
+from os import path
+from safetensors.torch import load_file
+from huggingface_hub import hf_hub_download
 
-    if USE_TORCH_COMPILE:
-        pipe.transformer = torch.compile(pipe.transformer, mode="reduce-overhead", fullgraph=True)
-        print("Model Compiled!")
+cache_path = path.join(path.dirname(path.abspath(__file__)), "models")
+os.environ["TRANSFORMERS_CACHE"] = cache_path
+os.environ["HF_HUB_CACHE"] = cache_path
+os.environ["HF_HOME"] = cache_path
 
+import gradio as gr
+import torch
+from diffusers import StableDiffusionXLPipeline, LCMScheduler
 
-def save_image(img):
-    unique_name = str(uuid.uuid4()) + ".png"
-    img.save(unique_name)
-    return unique_name
+# from scheduling_tcd import TCDScheduler
 
+torch.backends.cuda.matmul.allow_tf32 = True
 
-def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
-    if randomize_seed:
-        seed = random.randint(0, MAX_SEED)
-    return seed
+class timer:
+    def __init__(self, method_name="timed process"):
+        self.method = method_name
 
-@spaces.GPU(duration=30)
-def generate(
-        prompt: str,
-        negative_prompt: str = "",
-        style: str = DEFAULT_STYLE_NAME,
-        use_negative_prompt: bool = False,
-        seed: int = 0,
-        width: int = 1024,
-        height: int = 1024,
-        inference_steps: int = 4,
-        randomize_seed: bool = False,
-        use_resolution_binning: bool = True,
-        progress=gr.Progress(track_tqdm=True),
-):
-    seed = int(randomize_seed_fn(seed, randomize_seed))
-    generator = torch.Generator().manual_seed(seed)
-    
-    if not use_negative_prompt:
-        negative_prompt = None  # type: ignore
-    prompt, negative_prompt = apply_style(style, prompt, negative_prompt)
+    def __enter__(self):
+        self.start = time.time()
+        print(f"{self.method} starts")
 
-    images = pipe(
-        prompt=prompt,
-        negative_prompt=negative_prompt,
-        width=width,
-        height=height,
-        guidance_scale=0,
-        num_inference_steps=inference_steps,
-        generator=generator,
-        num_images_per_prompt=NUM_IMAGES_PER_PROMPT,
-        use_resolution_binning=use_resolution_binning,
-        output_type="pil",
-    ).images
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        end = time.time()
+        print(f"{self.method} took {str(round(end - self.start, 2))}s")
 
-    image_paths = [save_image(img) for img in images]
-    print(image_paths)
-    return image_paths, seed
+if not path.exists(cache_path):
+    os.makedirs(cache_path, exist_ok=True)
 
-
-examples = [
-    "A Monkey with a happy face in the Sahara desert.",
-    "Eiffel Tower was Made up of ICE.",
-    "Color photo of a corgi made of transparent glass, standing on the riverside in Yosemite National Park.",
-    "A close-up photo of a woman. She wore a blue coat with a gray dress underneath and has blue eyes.",
-    "A litter of golden retriever puppies playing in the snow. Their heads pop out of the snow, covered in.",
-    "an astronaut sitting in a diner, eating fries, cinematic, analog film",
-]
+pipe = StableDiffusionXLPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.bfloat16)
+pipe.to(device="cuda", dtype=torch.bfloat16)
+unet_state = load_file(hf_hub_download("ByteDance/Hyper-SD", "Hyper-SDXL-1step-Unet.safetensors"), device="cuda")
+pipe.unet.load_state_dict(unet_state)
+pipe.scheduler = LCMScheduler.from_config(pipe.scheduler.config, timestep_spacing ="trailing")
 
 with gr.Blocks() as demo:
     gr.Markdown(DESCRIPTION)
@@ -188,29 +58,6 @@ with gr.Blocks() as demo:
     with gr.Accordion("Advanced options", open=False):
         with gr.Group():
             with gr.Row():
-                use_negative_prompt = gr.Checkbox(label="Use negative prompt", value=False, visible=True)
-                negative_prompt = gr.Text(
-                label="Negative prompt",
-                max_lines=1,
-                placeholder="Enter a negative prompt",
-                visible=True,
-            )
-              
-#                num_imgs = gr.Slider(
- #                   label="Num Images",
-  #                  minimum=1,
-   #                 maximum=8,
-    #                step=1,
-     #               value=1,
-      #          )
-            style_selection = gr.Radio(
-                show_label=True,
-                container=True,
-                interactive=True,
-                choices=STYLE_NAMES,
-                value=DEFAULT_STYLE_NAME,
-                label="Image Style",
-            )
             seed = gr.Slider(
                 label="Seed",
                 minimum=0,
@@ -223,64 +70,69 @@ with gr.Blocks() as demo:
                 width = gr.Slider(
                     label="Width",
                     minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
+                    maximum=8192,
                     step=32,
-                    value=1024,
+                    value=2048,
                 )
                 height = gr.Slider(
                     label="Height",
                     minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
+                    maximum=8192,
                     step=32,
-                    value=1024,
+                    value=2048,
                 )
-        with gr.Row():
-            inference_steps = gr.Slider(
-                label="Steps",
-                minimum=4,
-                maximum=20,
-                step=1,
-                value=4,
-            )
 
-    gr.Examples(
-        examples=examples,
-        inputs=prompt,
-        outputs=[result, seed],
-        fn=generate,
-        cache_examples=CACHE_EXAMPLES,
-    )
+            def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
+            if randomize_seed:
+            seed = random.randint(0, MAX_SEED)
+            return seed
+            
+            @spaces.GPU(duration=10)
+            def process_image( height, width, prompt, seed, randomize_seed):
+                global pipe
+                with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16), timer("inference"):
+                    return pipe(
+                        prompt=str,,
+                        generator=torch.Generator().manual_seed(int(seed)),
+                        num_inference_steps=1,
+                        guidance_scale=0.,
+                        height=int(height),
+                        width=int(width),
+                        timesteps=[800],
+                        randomize_seed: bool = False,
+                        use_resolution_binning: bool = True,
+                        progress=gr.Progress(track_tqdm=True),
+                        ).images
+
+                        seed = int(randomize_seed_fn(seed, randomize_seed))
+                        generator = torch.Generator().manual_seed(seed)
+    
+            reactive_controls = [ height, width, prompt, seed, randomize_seed]
 
-    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,
-            style_selection,
-            use_negative_prompt,
-            # num_imgs,
-            seed,
-            width,
-            height,
-            inference_steps,
-            randomize_seed,
-        ],
-        outputs=[result, seed],
-        api_name="run",
-    )
+            btn.click(process_image, inputs=reactive_controls, outputs=[output])
 
 if __name__ == "__main__":
-    demo.queue(max_size=20).launch()
-    # demo.queue(max_size=20).launch(server_name="0.0.0.0", server_port=11900, debug=True)
+    demo.launch()
+
+
+DESCRIPTION = """ # Instant Image
+        ### Super fast text to Image Generator.
+        ### <span style='color: red;'>You may change the steps from 4 to 8, if you didn't get satisfied results.
+        ### First Image processing takes time then images generate faster.
+        """
+if not torch.cuda.is_available():
+    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"
+
+
+CACHE_EXAMPLES = torch.cuda.is_available() and os.getenv("CACHE_EXAMPLES", "1") == "1"
+        
+examples = [
+    "A Monkey with a happy face in the Sahara desert.",
+    "Eiffel Tower was Made up of ICE.",
+    "Color photo of a corgi made of transparent glass, standing on the riverside in Yosemite National Park.",
+    "A close-up photo of a woman. She wore a blue coat with a gray dress underneath and has blue eyes.",
+    "A litter of golden retriever puppies playing in the snow. Their heads pop out of the snow, covered in.",
+    "an astronaut sitting in a diner, eating fries, cinematic, analog film",
+]
+