Handling New Data

Models can be trained once in a set of data and can then be used to make predictions. Such static models can be very useful, but it is often the case that we want our model to continuously learn from new data—and to continuously get better as it does so.

In this section, we will discuss two strategies on how to re-train a deep learning model and how to implement them in Python.

Separating Data and Model

When building a deep learning application, the two most important areas are data and model. From an architectural point of view, we suggest that these two areas be separate. We believe that is a good suggestion because each of these areas include functions inherently separated from each other. Data is often required to be collected, ...

Get Beginning Application Development with TensorFlow and Keras now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.