Hands-On Machine Learning with TensorFlow.js

Supervised learning

A machine learning problem whose training dataset is composed of a feature vector and a target vector (also known as a label) is called supervised learning. Supervised learning is a type of machine learning that requires us to specify the target value explicitly. The training dataset must contain the target vector, as well as feature vectors. As shown in the following diagram, the algorithm generates the model in the training phase with the given training dataset, including the feature vector and target values. In the prediction phase, the model is asked to estimate the target value that corresponds to the given feature vector:

Roughly, there are three types of machine learning application processes: supervised learning, ...