When transfer learning should be used

Transfer learning works really well when you have limited data and when a network exists that solves a similar problem. You can use transfer learning to bring state-of-the art networks and giant volumes of data to an otherwise small problem. So, when should you use transfer learning? Anytime you can! But, there are two stipulations that I'd like you to think about first. We will discuss them in the following sections.

Limited data

The question I'm most often asked when it comes to computer vision and transfer learning is: How many images do I have to have? It's a difficult question to answer because, as we will see in the next section, more is usually better. A better question might be: How few images can I use to solve my business problem adequately?

So, just how limited can our dataset be? While far from scientific, I have built useful models using as few as 2,000 images for binary classification tasks. Simpler tasks and more diverse image sets typically...