Serving tensorflow model with Flask

The aim of this project is to provide a simple way you can quickly serve your tensorflow model.

Note: This code is not optimized for production enviroment.

Am using the model trained on the ImageNet Large Visual Recognition Challenge dataset which is described in this codelab.

Requirement

• Python >=2.7 or >=3.4
• TensorFlow >=1.0
• cv2 for image preprocessing [Feel free to use other libraries]
• Flask

Classifications

• daisy
• dandelion
• roses
• sunflowers
• tulip

How to run

$python app.py

Then open localhost:5000