python code to link front end and backend

app4.py

+from flask import Flask, render_template, request
+import tensorflow as tf
+from tensorflow.keras.preprocessing import image
+import numpy as np
+from PIL import Image
+import io
+from tensorflow.keras.preprocessing.image import ImageDataGenerator
+
+app = Flask(__name__)
+
+# Define a global variable to hold the loaded model
+model = None
+
+# Define a global session to ensure it's reused
+session = None
+
+# Define data directories
+data_dir = r'C:\Users\Maneka Wijesundara\Desktop\catResearch\cat_v1'
+
+# Define image size and batch size
+img_size = (150, 150)
+batch_size = 32
+
+# Define the allowed breeds
+allowed_breeds = ["siamese", "bengal", "main_coon", "ragdoll", "domestic_shorthair"]
+
+# Data augmentation and preprocessing
+datagen = ImageDataGenerator(
+    rescale=1.0/255.0,
+    rotation_range=20,
+    width_shift_range=0.2,
+    height_shift_range=0.2,
+    horizontal_flip=True,
+    validation_split=0.2
+)
+
+# Load and split the data
+train_generator = datagen.flow_from_directory(
+    data_dir,
+    target_size=img_size,
+    batch_size=batch_size,
+    class_mode='categorical',
+    subset='training'
+)
+
+def load_model():
+    global model, session
+    if model is None:
+        model = tf.keras.models.load_model('cat_breed_classifier.h5')  # Replace 'my_model.h5' with the actual model path
+        session = tf.keras.backend.get_session()
+
+def preprocess_image(file):
+    img = Image.open(io.BytesIO(file.read()))
+    img = img.resize((150, 150))  # Resize to 150x150 pixels
+    img = np.array(img)
+    img = np.expand_dims(img, axis=0)
+    img = img / 255.0  # Normalize pixel values to [0, 1]
+    return img
+
+@app.route("/", methods=["GET", "POST"])
+def index():
+    predicted_breed = ""
+    cat_image = None
+
+    if request.method == "POST":
+        file = request.files["file"]
+
+        if file:
+            load_model()  # Load the model
+
+            img = preprocess_image(file)
+
+            with session.as_default():
+                with session.graph.as_default():
+                    predictions = model.predict(img)
+
+            class_indices = train_generator.class_indices
+            predicted_class = np.argmax(predictions)
+
+            for breed, idx in class_indices.items():
+                if idx == predicted_class:
+                    predicted_breed = breed
+                    break
+            else:
+                predicted_breed = "Unknown"
+
+            if predicted_breed not in allowed_breeds:
+                message = "Breed not recognized. The model is trained to identify Siamese, Bengal, Main Coon, Ragdoll, and Domestic Short Hair."
+                return render_template("index.html", predicted_breed=message, cat_image=cat_image)
+
+            # Save the image for rendering in the template
+            cat_image = file.read()
+
+    return render_template("index.html", predicted_breed=predicted_breed, cat_image=cat_image)
+
+if __name__ == "__main__":
+    app.run(debug=True, port=8080)