unit 5.3 - Fine tuning a pre-trained Neural Network

In many circumstances, you will want to create a neural network based on a small set of your own data. However, training a neural network from scratch can be computationally expensive and time-consuming.

In this unit we will learn how to fine-tune a pre-trained neural network. This is useful if you have some data that is similar to the data that the pre-trained network was trained on, but not exactly the same. This technique uses a pre-trained neural network as a feature extractor, and then trains a new classifier on top of the extracted features.

An example of fine-tuning

See this example as a starting point.