app.py

app.py

@@ -0,0 +1,72 @@
+import os
+import numpy as np
+import pandas as pd
+from flask import Flask, request, jsonify
+from tensorflow.keras.models import load_model
+from tensorflow.keras.preprocessing.image import load_img, img_to_array
+from tensorflow.keras.applications.resnet50 import preprocess_input
+from sklearn.preprocessing import StandardScaler
+
+# Flask app setup
+app = Flask(_name_)
+
+# Load the model
+MODEL_PATH = "fine_tuned_lake_event_prediction_model_with_xgb.keras"
+model = load_model(MODEL_PATH)
+
+# Parameters
+time_steps_image = 4
+target_img_size = (224, 224)
+
+# Preprocessing functions
+def preprocess_sensor_data(sensor_data):
+    """Preprocess sensor data by normalizing."""
+    scaler = StandardScaler()
+    return scaler.fit_transform(sensor_data)
+
+def preprocess_image(image_path):
+    """Preprocess a single image."""
+    try:
+        image = load_img(image_path, target_size=target_img_size)
+        image_array = img_to_array(image)
+        return preprocess_input(image_array)
+    except Exception as e:
+        print(f"Error processing image {image_path}: {e}")
+        return np.zeros((224, 224, 3), dtype=np.float32)
+
+def preprocess_image_sequences(image_paths):
+    """Preprocess a sequence of images."""
+    images = [preprocess_image(img_path) for img_path in image_paths]
+    return np.expand_dims(np.array(images), axis=0)
+
+# Routes
+@app.route('/predict', methods=['POST'])
+def predict():
+    """Predict based on sensor data and image sequences."""
+    try:
+        # Extract JSON payload
+        data = request.json
+
+        # Sensor data preprocessing
+        sensor_data = np.array(data['sensor_data']).reshape(1, -1)
+        sensor_data = preprocess_sensor_data(sensor_data)
+
+        # Image data preprocessing
+        image_paths = data['image_paths']
+        if len(image_paths) != time_steps_image:
+            return jsonify({"error": f"Exactly {time_steps_image} image paths are required."}), 400
+
+        image_sequences = preprocess_image_sequences(image_paths)
+
+        # Make prediction
+        prediction = model.predict([sensor_data, image_sequences])
+        result = float(prediction[0][0])
+
+        return jsonify({"prediction": result})
+
+    except Exception as e:
+        return jsonify({"error": str(e)}), 500
+
+# Main entry point
+if _name_ == "_main_":
+    app.run(debug=True)