Update app.py

app.py

@@ -1,56 +1,74 @@
 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
+import re
 
+# Устройство и параметры загрузки модели
 device = "cuda" if torch.cuda.is_available() else "cpu"
-model_repo_id = "stabilityai/sdxl-turbo"  # Replace to the model you would like to use
+torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+
+# Регулярное выражение для проверки корректности модели
+VALID_REPO_ID_REGEX = re.compile(r"^[a-zA-Z0-9._\-]+/[a-zA-Z0-9._\-]+$")
+
+def is_valid_repo_id(repo_id):
+    return bool(VALID_REPO_ID_REGEX.match(repo_id)) and not repo_id.endswith(('-', '.'))
 
-if torch.cuda.is_available():
-    torch_dtype = torch.float16
-else:
-    torch_dtype = torch.float32
+# Изначально загружаем модель по умолчанию
+model_repo_id = "CompVis/stable-diffusion-v1-4"
+pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype).to(device)
 
-pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
-pipe = pipe.to(device)
+# Загружаем веса LoRA
+pipe.load_lora_weights("AnastasiaSh/sticker-cat-lora3")
 
 MAX_SEED = np.iinfo(np.int32).max
 MAX_IMAGE_SIZE = 1024
 
-
-# @spaces.GPU #[uncomment to use ZeroGPU]
 def infer(
+    model,
     prompt,
     negative_prompt,
     seed,
-    randomize_seed,
     width,
     height,
     guidance_scale,
     num_inference_steps,
     progress=gr.Progress(track_tqdm=True),
 ):
-    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]
+    global model_repo_id, pipe
+
+    # Проверяем и загружаем новую модель, если она изменена
+    if model != model_repo_id:
+        if not is_valid_repo_id(model):
+            raise gr.Error(f"Некорректный идентификатор модели: '{model}'. Проверьте название.")
+        try:
+            new_pipe = DiffusionPipeline.from_pretrained(model, torch_dtype=torch_dtype).to(device)
+            # Повторно загружаем LoRA для новой модели
+            new_pipe.load_lora_weights("AnastasiaSh/sticker-cat-lora3")
+            pipe = new_pipe
+            model_repo_id = model
+        except Exception as e:
+            raise gr.Error(f"Не удалось загрузить модель '{model}'.\nОшибка: {e}")
+
+    # Генератор случайных чисел для детерминированности
+    generator = torch.Generator(device=device).manual_seed(seed)
+
+    # Генерация изображения
+    try:
+        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]
+    except Exception as e:
+        raise gr.Error(f"Ошибка при генерации изображения: {e}")
 
     return image, seed
 
-
 examples = [
     "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
     "An astronaut riding a green horse",
@@ -66,89 +84,89 @@ css = """
 
 with gr.Blocks(css=css) as demo:
     with gr.Column(elem_id="col-container"):
-        gr.Markdown(" # Text-to-Image Gradio Template")
-
-        with gr.Row():
-            prompt = gr.Text(
-                label="Prompt",
-                show_label=False,
-                max_lines=1,
-                placeholder="Enter your prompt",
-                container=False,
-            )
-
-            run_button = gr.Button("Run", scale=0, variant="primary")
-
+        gr.Markdown("# Text-to-Image App")
+
+        model = gr.Textbox(
+            label="Model",
+            value="CompVis/stable-diffusion-v1-4",  # Значение по умолчанию
+            interactive=True
+        )
+
+        prompt = gr.Text(
+            label="Prompt",
+            show_label=False,
+            max_lines=1,
+            placeholder="Enter your prompt",
+            container=False,
+        )
+
+        negative_prompt = gr.Text(
+            label="Negative prompt",
+            max_lines=1,
+            placeholder="Enter a negative prompt",
+            visible=True,
+        )
+
+        seed = gr.Slider(
+            label="Seed",
+            minimum=0,
+            maximum=MAX_SEED,
+            step=1,
+            value=42,
+        )
+
+        guidance_scale = gr.Slider(
+            label="Guidance scale",
+            minimum=0.0,
+            maximum=10.0,
+            step=0.1,
+            value=7.0,
+        )
+
+        num_inference_steps = gr.Slider(
+            label="Number of inference steps",
+            minimum=1,
+            maximum=50,
+            step=1,
+            value=20,
+        )
+
+        run_button = gr.Button("Run", variant="primary")
         result = gr.Image(label="Result", show_label=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,
-            )
-
-            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
+                    value=512,
                 )
-
                 height = gr.Slider(
                     label="Height",
                     minimum=256,
                     maximum=MAX_IMAGE_SIZE,
                     step=32,
-                    value=1024,  # Replace with defaults that work for your model
-                )
-
-            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
+                    value=512,
                 )
 
         gr.Examples(examples=examples, inputs=[prompt])
-    gr.on(
-        triggers=[run_button.click, prompt.submit],
-        fn=infer,
-        inputs=[
-            prompt,
-            negative_prompt,
-            seed,
-            randomize_seed,
-            width,
-            height,
-            guidance_scale,
-            num_inference_steps,
-        ],
-        outputs=[result, seed],
-    )
+
+        run_button.click(
+            infer,
+            inputs=[
+                model,
+                prompt,
+                negative_prompt,
+                seed,
+                width,
+                height,
+                guidance_scale,
+                num_inference_steps,
+            ],
+            outputs=[result, seed],
+        )
 
 if __name__ == "__main__":
     demo.launch()