ed

app.py

@@ -0,0 +1,75 @@
+from flask import Flask, request, render_template, redirect, url_for
+from PIL import Image
+from tensorflow.keras.models import load_model
+import numpy as np
+import os
+import logging
+import traceback
+
+app = Flask(__name__)
+
+# Configure detailed logging
+logging.basicConfig(
+    level=logging.DEBUG,
+    format='%(asctime)s - %(levelname)s - %(message)s'
+)
+logger = logging.getLogger(__name__)
+
+# Load models at startup
+models = {}
+try:
+    model_dir = 'saved_models'
+    if not os.path.exists(model_dir):
+        raise Exception(f"Directory {model_dir} not found")
+
+    for model_file in os.listdir(model_dir):
+        if model_file.endswith('.keras'):
+            model_path = os.path.join(model_dir, model_file)
+            models[model_file] = load_model(model_path)
+            logger.info(f"Loaded model: {model_file}")
+except Exception as e:
+    logger.error(f"Error loading models: {str(e)}")
+    logger.error(traceback.format_exc())
+
+@app.route('/')
+def home():
+    return render_template('index.html')
+
+@app.route('/predict', methods=['GET', 'POST'])
+def predict():
+    if request.method == 'GET':
+        return redirect(url_for('home'))
+
+    try:
+        if 'image' not in request.files:
+            raise ValueError("No image file provided")
+
+        file = request.files['image']
+        if file.filename == '':
+            raise ValueError("No selected file")
+
+        img = Image.open(file)
+        img = img.convert('RGB')  # Ensure image is in RGB format
+        img = img.resize((32, 32))
+        img_array = np.array(img) / 255.0
+        img_array = np.expand_dims(img_array, axis=0)
+
+        results = []
+        for model_name, model in models.items():
+            pred = model.predict(img_array)
+            class_idx = np.argmax(pred[0])
+            confidence = float(pred[0][class_idx])
+            results.append({
+                'model': model_name,
+                'class': int(class_idx),
+                'confidence': f"{confidence:.4f}"
+            })
+
+        return render_template('index.html', results=results)
+
+    except Exception as e:
+        logger.error(f"Error during prediction: {str(e)}")
+        return render_template('index.html', error=str(e))
+
+if __name__ == '__main__':
+    app.run(port=5001, debug=True)