Types of problems TPOT can solve
The TPOT library was designed as a go-to tool for automating machine learning tasks; hence, it should be able to handle with ease anything you throw at it. We will start using TPOT in a practical sense soon. Chapter 3, Exploring before Regression, shows how to use the library to handle practical tasks with many examples, and the following chapters focus on other types of tasks.
In general, TPOT can be used to handle the following types of tasks:
- Regression: Where the target variable is continuous, such as age, height, weight, score, or price. Refer to Chapter 1, Machine Learning and the Idea of Automation, for a brief overview of regression.
- Classification: Where the target variable is categorical, such as sold/did not sell, churn/did not churn, or yes/no. Refer to Chapter 1, Machine Learning and the Idea of Automation, for a brief overview of classification.
- Parallel training: TPOT can handle the training of machine learning models...