Serve trained Tensorflow model with REST API using Flask?

Serving Trained TensorFlow Model with REST API using Flask

Introduction

This article guides you through the process of deploying a trained TensorFlow model as a REST API using Flask, a popular Python web framework. This allows you to seamlessly integrate your machine learning models into web applications or other systems.

Steps

1. Install Required Libraries

pip install tensorflow flask

2. Prepare Your Trained Model
• Ensure your TensorFlow model is trained and saved.
• You can use either a SavedModel or a Keras model.
3. Create the Flask Application

from flask import Flask, request, jsonify import tensorflow as tf app = Flask(__name__) # Load your trained model model = tf.keras.models.load_model('path/to/your/model') @app.route('/', methods=['GET']) def index(): return "Welcome to the TensorFlow Model API" @app.route('/predict', methods=['POST']) def predict(): data = request.get_json() # Preprocess the data # ... prediction = model.predict(data) # Postprocess the prediction # ... return jsonify({'prediction': prediction.tolist()}) if __name__ == '__main__': app.run(debug=True) 

4. Test the API
• Run the Flask application: python app.py
• Use a tool like Postman or curl to send a POST request to http://localhost:5000/predict with the input data in JSON format.

Example Code: Image Classification

from flask import Flask, request, jsonify from PIL import Image import numpy as np import tensorflow as tf app = Flask(__name__) # Load your trained model model = tf.keras.models.load_model('path/to/your/model') @app.route('/predict', methods=['POST']) def predict(): # Get the image from the request image_data = request.files['image'].read() image = Image.open(io.BytesIO(image_data)).convert('RGB') image = image.resize((224, 224)) image = np.array(image) / 255.0 image = np.expand_dims(image, axis=0) # Make a prediction prediction = model.predict(image) class_names = ['cat', 'dog'] # Example class names predicted_class = class_names[np.argmax(prediction)] # Return the prediction as a JSON response return jsonify({'predicted_class': predicted_class}) if __name__ == '__main__': app.run(debug=True) 

Output

 { "predicted_class": "cat" } 

Deployment

• Local Deployment: Run the Flask app locally.
• Cloud Deployment: Use services like AWS Lambda, Google Cloud Functions, or Heroku to deploy your API to the cloud.
• Docker: Containerize your app and deploy it to a Docker container.

Conclusion

By following these steps, you can effectively serve a trained TensorFlow model through a REST API using Flask. This enables you to leverage your models in various applications and enhance the functionality of your web projects.

Post Views: 8