Transfer Learning
In this section, we will look at what Transfer Learning is and how it is going to be useful for us as we build real-time object detection. We divide this section into the following parts:
What is Transfer Learning?
What is a pre-trained model?
Why should we use a pre-trained model?
How can we use the pre-trained model?
Let's start with the first question.
What is Transfer Learning?
We will be looking at the intuition behind Transfer Learning first and, then, we will cover its technical definition. Let me explain this concept through a simple teacher-student analogy. A teacher has many years of experience in teaching certain specific topics or subjects. Whatever information the teacher has, they deliver it to their students. So, the process of teaching is all about transferring knowledge from the teacher to the student. You can refer to the following figure:
Figure 9.3: An overview of Transfer Learning
Now, remember this analogy; we will apply it to neural networks. When we train...
