We have completed our deep learning model and stored its structure in the model.json file and the weights for the model in the weights.h5 file. We are now ready to wrap the model data in an API so that we can expose the model to web-based calls via the GET or POST methods. Here, we will be discussing the POST method. Let's begin with the required setup on the server.
Creating a Flask API to work with server-side Python
Setting up the environment
In the server, we will require the Flask module—which will be service requests—which in turn will be running code that requires Keras (and so, TensorFlow), NumPy, and many other modules. In order to quickly set up the environment for our project, we follow these...
