Neural Network Projects with Python

Naive one-shot prediction – Euclidean distance between two vectors

Before we dive into how neural networks can be used for one-shot learning, let's look at one naive approach.

Given the true image and a test image, one naive approach for a one-shot prediction is to simply measure the difference between the two images. As we have already seen, all images are simply three-dimensional vectors. We know that the Euclidean distance provides a mathematical formulation of the difference between two vectors. To refresh your memory, the Euclidean distance between two vectors is shown in the following diagram:

Measuring the Euclidean distance between ...