Creating Effective Datasets for Machine Learning

A label is an identifier used to represent classes, and it is typically represented as integers starting with 0.

Features in machine learning models are always represented as numbers.

Classification involves putting things into discrete categories, or classes, like dog breed, flower type, and digit.

The goal of training the model is to learn a relationship between the input features and the output label.

If the training data is not a good representation of the data the model will receive when it is used, we can expect our model to perform well.