1. from flask import Flask, request, jsonify, render_template
2. import tensorflow as tf
3. from tensorflow.keras.models import load_model
4. import numpy as np
5. from PIL import Image
6. import io
7.  
8. app = Flask(name)
9.  
10. # Load your pre-trained model
11. model = load_model('D:Versity8TH SEMESTERbrain_tumorBrain-Tumor-Segmentation-in-TensorFlow-2.0-mainfiles/model.h5')
12.  
13. def preprocess_image(image):
14.     # Preprocess the image as required by your model (resize, normalize, etc.)
15.     image = image.resize((256, 256))  # Example size
16.     image = np.array(image) / 255.0   # Normalize
17.     image = np.expand_dims(image, axis=0)  # Add batch dimension
18.     return image
19.  
20. @app.route('/')
21. def home():
22.     return render_template('index.html')  # Serve your HTML page
23.  
24. @app.route('/predict', methods=['POST'])
25. def predict():
26.     if 'file' not in request.files:
27.         return "No file part"
28.     file = request.files['file']
29.     if file.filename == '':
30.         return "No selected file"
31.     if file:
32.         img = Image.open(io.BytesIO(file.read()))
33.         img = preprocess_image(img)
34.         prediction = model.predict(img)
35.         result = np.argmax(prediction, axis=-1)
36.         result = result[0].tolist()  # Convert the prediction to a list format
37.         return jsonify(result)
38.  
39. if name == 'main':
40.     app.run(debug=True)