Create app.py

app.py

@@ -0,0 +1,81 @@
+import os
+import sys
+import tensorflow as tf
+import numpy as np
+from tensorflow.keras import layers, models
+import requests
+from PIL import Image
+from io import BytesIO
+import gradio as gr
+
+class FoodQualityAI:
+    def __init__(self):
+        self.model = None
+        self.image_size = (256, 256)
+
+    def test(self, image_url):
+        if self.model is None:
+            raise Exception("Model not loaded. Please load a model before testing.")
+
+        try:
+            response = requests.get(image_url)
+            response.raise_for_status()
+        except Exception as ex:
+            raise Exception("Error downloading the image: " + str(ex))
+            
+        try:
+            img = Image.open(BytesIO(response.content)).convert("RGB")
+        except Exception as e:
+            raise Exception("Error processing the image: " + str(e))
+
+        img = img.resize(self.image_size)
+        img_array = np.array(img) / 255.0
+        img_array = np.expand_dims(img_array, axis=0)
+
+        prediction = self.model.predict(img_array)
+        
+        class_names = [d for d in os.listdir(self.train_dir) 
+                       if os.path.isdir(os.path.join(self.train_dir, d))]
+        if not class_names:
+            num_classes = prediction.shape[1]
+            class_names = [f"class_{i}" for i in range(num_classes)]
+            
+        predicted_idx = np.argmax(prediction)
+        predicted_class = class_names[predicted_idx]
+        confidence = float(np.max(prediction))
+
+        return predicted_class, confidence
+
+    def load(self, filename='food_quality_model.h5'):
+        try:
+            self.model = models.load_model(filename)
+        except Exception as e:
+            raise Exception(f"Не удалсь загрузить модель ИИ {filename}: {str(e)}")
+
+
+food_quality_ai = FoodQualityAI()
+
+try:
+    food_quality_ai.load('food_quality_model.h5')
+    print("Model loaded successfully.")
+except Exception as e:
+    print("Error loading model:", e)
+    sys.exit(1)
+
+def classify_image_url(image_url):
+    try:
+        predicted_class, confidence = food_quality_ai.test(image_url)
+        return f"Модель думает что это: {predicted_class}\nС вероятностю в: {confidence:.2f}"
+    except Exception as err:
+        return f"Ошибка: {err}"
+
+iface = gr.Interface(
+    fn=classify_image_url,
+    inputs=gr.Textbox(label="Ссылка на тестовую картинку", placeholder="Введите ссылку..."),
+    outputs=gr.Textbox(label="Мнение модели"),
+    title="ИИ для определения состояния еды.",
+    description="Вставьте ссылку чтобы проверить как работает модель."
+)
+
+if __name__ == "__main__":
+    iface.launch()