Update app.py

app.py

@@ -3,8 +3,13 @@ import cv2
 import numpy as np
 from flask import Flask, request, render_template, jsonify
 from werkzeug.utils import secure_filename
+
+# Suppress TensorFlow warnings
+os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
+
 import tensorflow as tf
-from tensorflow.keras.models import load_model
+# Use standalone Keras (compatible with old models)
+from keras.models import load_model
 
 app = Flask(__name__)
 app.config['MAX_CONTENT_LENGTH'] = 16 * 1024 * 1024  # 16MB max file size
@@ -13,26 +18,17 @@ app.config['UPLOAD_FOLDER'] = 'uploads'
 # Create uploads folder if it doesn't exist
 os.makedirs(app.config['UPLOAD_FOLDER'], exist_ok=True)
 
-# Load the model with compatibility handling
+# Load the model with old Keras 2.4.3
 print("Loading model...")
 import warnings
 warnings.filterwarnings('ignore')
 
-# Set TensorFlow to use less memory
-os.environ['TF_FORCE_GPU_ALLOW_GROWTH'] = 'true'
-
-# Try loading with h5py directly for better compatibility
-import h5py
-
-# Load model - TensorFlow 2.15 should handle it
 try:
-    # Disable safe mode for compatibility
-    model = tf.keras.models.load_model('cancer_model.h5', compile=False, safe_mode=False)
-    print("✓ Model loaded successfully with TensorFlow 2.15!")
+    # Use standalone Keras which supports old batch_shape parameter
+    model = load_model('cancer_model.h5', compile=False)
+    print("✓ Model loaded successfully with Keras 2.4.3!")
 except Exception as e:
-    print(f"Error loading model: {e}")
-    print("\n⚠️ If loading fails, the model needs to be re-saved.")
-    print("Please run the convert_model.py script provided.")
+    print(f"❌ Error loading model: {e}")
     raise
 
 def resize_with_padding(img, target_size):