Update app.py

app.py

@@ -27,7 +27,6 @@ MAX_IMAGE_SIZE = int(os.getenv("MAX_IMAGE_SIZE", "2048"))
 USE_TORCH_COMPILE = os.getenv("USE_TORCH_COMPILE") == "1"
 ENABLE_CPU_OFFLOAD = os.getenv("ENABLE_CPU_OFFLOAD") == "1"
 OUTPUT_DIR = os.getenv("OUTPUT_DIR", "./outputs")
-THUMBNAIL_SIZE = (128, 128)  # Size for thumbnails
 
 MODEL = os.getenv(
     "MODEL",
@@ -39,11 +38,33 @@ torch.backends.cudnn.benchmark = False
 
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
-# Store the generation history
-generation_history = []
+# Add a new global variable to store the image history
+image_history = []
 
 def load_pipeline(model_name):
-    # ... (rest of the function remains the same)
+    vae = AutoencoderKL.from_pretrained(
+        "madebyollin/sdxl-vae-fp16-fix",
+        torch_dtype=torch.float16,
+    )
+    pipeline = (
+        StableDiffusionXLPipeline.from_single_file
+        if MODEL.endswith(".safetensors")
+        else StableDiffusionXLPipeline.from_pretrained
+    )
+
+    pipe = pipeline(
+        model_name,
+        vae=vae,
+        torch_dtype=torch.float16,
+        custom_pipeline="lpw_stable_diffusion_xl",
+        use_safetensors=True,
+        add_watermarker=False,
+        use_auth_token=HF_TOKEN,
+        variant="fp16",
+    )
+
+    pipe.to(device)
+    return pipe
 
 @spaces.GPU
 def generate(
@@ -61,29 +82,95 @@ def generate(
     upscale_by: float = 1.5,
     progress=gr.Progress(track_tqdm=True),
 ) -> Image:
-    # ... (rest of the function remains the same)
+    generator = utils.seed_everything(seed)
+
+    width, height = utils.aspect_ratio_handler(
+        aspect_ratio_selector,
+        custom_width,
+        custom_height,
+    )
+
+    width, height = utils.preprocess_image_dimensions(width, height)
+
+    backup_scheduler = pipe.scheduler
+    pipe.scheduler = utils.get_scheduler(pipe.scheduler.config, sampler)
+
+    if use_upscaler:
+        upscaler_pipe = StableDiffusionXLImg2ImgPipeline(**pipe.components)
+    metadata = {
+        "prompt": prompt,
+        "negative_prompt": negative_prompt,
+        "resolution": f"{width} x {height}",
+        "guidance_scale": guidance_scale,
+        "num_inference_steps": num_inference_steps,
+        "seed": seed,
+        "sampler": sampler,
+    }
+
+    if use_upscaler:
+        new_width = int(width * upscale_by)
+        new_height = int(height * upscale_by)
+        metadata["use_upscaler"] = {
+            "upscale_method": "nearest-exact",
+            "upscaler_strength": upscaler_strength,
+            "upscale_by": upscale_by,
+            "new_resolution": f"{new_width} x {new_height}",
+        }
+    else:
+        metadata["use_upscaler"] = None
+    logger.info(json.dumps(metadata, indent=4))
 
     try:
-        # ... (existing code for image generation)
+        if use_upscaler:
+            latents = pipe(
+                prompt=prompt,
+                negative_prompt=negative_prompt,
+                width=width,
+                height=height,
+                guidance_scale=guidance_scale,
+                num_inference_steps=num_inference_steps,
+                generator=generator,
+                output_type="latent",
+            ).images
+            upscaled_latents = utils.upscale(latents, "nearest-exact", upscale_by)
+            images = upscaler_pipe(
+                prompt=prompt,
+                negative_prompt=negative_prompt,
+                image=upscaled_latents,
+                guidance_scale=guidance_scale,
+                num_inference_steps=num_inference_steps,
+                strength=upscaler_strength,
+                generator=generator,
+                output_type="pil",
+            ).images
+        else:
+            images = pipe(
+                prompt=prompt,
+                negative_prompt=negative_prompt,
+                width=width,
+                height=height,
+                guidance_scale=guidance_scale,
+                num_inference_steps=num_inference_steps,
+                generator=generator,
+                output_type="pil",
+            ).images
 
-        if images:
-            # Create thumbnail
-            thumbnail = images[0].copy()
-            thumbnail.thumbnail(THUMBNAIL_SIZE)
-            
-            # Add to generation history
-            generation_history.append({
+        if images and IS_COLAB:
+            for image in images:
+                filepath = utils.save_image(image, metadata, OUTPUT_DIR)
+                logger.info(f"Image saved as {filepath} with metadata")
+
+        # Add the generated image and metadata to the history
+        for image in images:
+            thumbnail = image.copy()
+            thumbnail.thumbnail((256, 256))
+            image_history.insert(0, {
+                "image": thumbnail,
                 "prompt": prompt,
-                "thumbnail": thumbnail,
                 "metadata": metadata
             })
 
-            if IS_COLAB:
-                for image in images:
-                    filepath = utils.save_image(image, metadata, OUTPUT_DIR)
-                    logger.info(f"Image saved as {filepath} with metadata")
-
-        return images, metadata, update_history()
+        return images, metadata, gr.update(value=image_history)
     except Exception as e:
         logger.exception(f"An error occurred: {e}")
         raise
@@ -93,19 +180,6 @@ def generate(
         pipe.scheduler = backup_scheduler
         utils.free_memory()
 
-def update_history():
-    history_html = "<div style='display: flex; flex-wrap: wrap;'>"
-    for item in reversed(generation_history[-10:]):  # Show last 10 entries
-        thumbnail_path = f"data:image/png;base64,{utils.image_to_base64(item['thumbnail'])}"
-        history_html += f"""
-        <div style='margin: 5px; text-align: center;'>
-            <img src='{thumbnail_path}' style='width: 100px; height: 100px; object-fit: cover;'>
-            <p style='font-size: 12px; margin: 5px 0;'>{item['prompt'][:50]}...</p>
-        </div>
-        """
-    history_html += "</div>"
-    return history_html
-
 if torch.cuda.is_available():
     pipe = load_pipeline(MODEL)
     logger.info("Loaded on Device!")
@@ -128,43 +202,133 @@ with gr.Blocks(css="style.css") as demo:
     )
     with gr.Group():
         with gr.Row():
-            prompt = gr.Text(
-                label="Prompt",
-                show_label=False,
-                max_lines=5,
-                placeholder="Enter your prompt",
-                container=False,
-            )
-            run_button = gr.Button(
-                "Generate", 
-                variant="primary", 
-                scale=0
-            )
-        result = gr.Gallery(
-            label="Result", 
-            columns=1, 
-            preview=True, 
-            show_label=False
-        )
-    
-    # Add the history display
-    history_display = gr.HTML(label="Generation History")
+            with gr.Column(scale=2):
+                prompt = gr.Text(
+                    label="Prompt",
+                    show_label=False,
+                    max_lines=5,
+                    placeholder="Enter your prompt",
+                    container=False,
+                )
+                run_button = gr.Button(
+                    "Generate", 
+                    variant="primary", 
+                    scale=0
+                )
+                result = gr.Gallery(
+                    label="Result", 
+                    columns=1, 
+                    preview=True, 
+                    show_label=False
+                )
+            
+            with gr.Column(scale=1):
+                history = gr.Gallery(
+                    label="Generation History",
+                    show_label=True,
+                    elem_id="history",
+                    columns=2,
+                    height=800,
+                )
 
     with gr.Accordion(label="Advanced Settings", open=False):
-        # ... (rest of the UI components remain the same)
+        negative_prompt = gr.Text(
+            label="Negative Prompt",
+            max_lines=5,
+            placeholder="Enter a negative prompt",
+            value=""
+        )
+        aspect_ratio_selector = gr.Radio(
+            label="Aspect Ratio",
+            choices=config.aspect_ratios,
+            value="1024 x 1024",
+            container=True,
+        )
+        with gr.Group(visible=False) as custom_resolution:
+            with gr.Row():
+                custom_width = gr.Slider(
+                    label="Width",
+                    minimum=MIN_IMAGE_SIZE,
+                    maximum=MAX_IMAGE_SIZE,
+                    step=8,
+                    value=1024,
+                )
+                custom_height = gr.Slider(
+                    label="Height",
+                    minimum=MIN_IMAGE_SIZE,
+                    maximum=MAX_IMAGE_SIZE,
+                    step=8,
+                    value=1024,
+                )
+        use_upscaler = gr.Checkbox(label="Use Upscaler", value=False)
+        with gr.Row() as upscaler_row:
+            upscaler_strength = gr.Slider(
+                label="Strength",
+                minimum=0,
+                maximum=1,
+                step=0.05,
+                value=0.55,
+                visible=False,
+            )
+            upscale_by = gr.Slider(
+                label="Upscale by",
+                minimum=1,
+                maximum=1.5,
+                step=0.1,
+                value=1.5,
+                visible=False,
+            )
 
+        sampler = gr.Dropdown(
+            label="Sampler",
+            choices=config.sampler_list,
+            interactive=True,
+            value="DPM++ 2M SDE Karras",
+        )
+        with gr.Row():
+            seed = gr.Slider(
+                label="Seed", minimum=0, maximum=utils.MAX_SEED, step=1, value=0
+            )
+            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+        with gr.Group():
+            with gr.Row():
+                guidance_scale = gr.Slider(
+                    label="Guidance scale",
+                    minimum=1,
+                    maximum=12,
+                    step=0.1,
+                    value=7.0,
+                )
+                num_inference_steps = gr.Slider(
+                    label="Number of inference steps",
+                    minimum=1,
+                    maximum=50,
+                    step=1,
+                    value=28,
+                )
     with gr.Accordion(label="Generation Parameters", open=False):
         gr_metadata = gr.JSON(label="Metadata", show_label=False)
-
     gr.Examples(
         examples=config.examples,
         inputs=prompt,
-        outputs=[result, gr_metadata, history_display],
+        outputs=[result, gr_metadata, history],
         fn=lambda *args, **kwargs: generate(*args, use_upscaler=True, **kwargs),
         cache_examples=CACHE_EXAMPLES,
     )
-
-    # ... (rest of the event handlers remain the same)
+    use_upscaler.change(
+        fn=lambda x: [gr.update(visible=x), gr.update(visible=x)],
+        inputs=use_upscaler,
+        outputs=[upscaler_strength, upscale_by],
+        queue=False,
+        api_name=False,
+    )
+    aspect_ratio_selector.change(
+        fn=lambda x: gr.update(visible=x == "Custom"),
+        inputs=aspect_ratio_selector,
+        outputs=custom_resolution,
+        queue=False,
+        api_name=False,
+    )
 
     inputs = [
         prompt,
@@ -190,7 +354,7 @@ with gr.Blocks(css="style.css") as demo:
     ).then(
         fn=generate,
         inputs=inputs,
-        outputs=[result, gr_metadata, history_display],
+        outputs=[result, gr_metadata, history],
         api_name="run",
     )
     negative_prompt.submit(
@@ -202,7 +366,7 @@ with gr.Blocks(css="style.css") as demo:
     ).then(
         fn=generate,
         inputs=inputs,
-        outputs=[result, gr_metadata, history_display],
+        outputs=[result, gr_metadata, history],
         api_name=False,
     )
     run_button.click(
@@ -214,7 +378,7 @@ with gr.Blocks(css="style.css") as demo:
     ).then(
         fn=generate,
         inputs=inputs,
-        outputs=[result, gr_metadata, history_display],
+        outputs=[result, gr_metadata, history],
         api_name=False,
     )