Machine Learning and AI Foundations: Value Estimations

Course details

Value estimation—one of the most common types of machine learning algorithms—can automatically estimate values by looking at related information. For example, a website can determine how much a house is worth based on the property's location and characteristics. In this project-based course, discover how to use machine learning to build a value estimation system that can deduce the value of a home. Follow Adam Geitgey as he walks through how to use sample data to build a machine learning model, and then use that model in your own programs. Although the project featured in this course focuses on real estate, you can use the same approach to solve any kind of value estimation problem with machine learning.

Topics include:
Setting up the development environment
Building a simple home value estimator
Finding the best weights automatically
Working with large data sets efficiently
Training a supervised machine learning model
Exploring a home value data set
Deciding how much data is needed
Preparing the features
Training the value estimator
Measuring accuracy with mean absolute error
Improving a system
Using the machine learning model to make predictions

Concepts

0. Introduction

01 - Welcome
02 - What you should know
03 - Using the exercise files
04 - Set up the development environment

1. What Is Machine Learning and Value Prediction

05 - What is machine learning
06 - Supervised machine learning for value prediction
07 - Build a simple home value estimator
08 - Find the best weights automatically
09 - Cool uses of value prediction

2. An Overview of Building a Machine Learning System

10 - Introduction to NumPy, scikit-learn, and pandas
11 - Think in vectors - How to work with large data sets efficiently
12 - The basic workflow for training a supervised machine learning model
13 - Gradient boosting - A versatile machine learning algorithm

3. Training Data

14 - Explore a home value data set
15 - Standard conventions for naming training data
16 - Decide how much data you need

4. Features

17 - Feature engineering
18 - Choose the best features for home value prediction
19 - Use as few features as possible - The curse of dimensionality

5. Coding Our System

20 - Prepare the features
21 - Training vs. testing data
22 - Train the value estimator
23 - Measure accuracy with mean absolute error

6. Improving Our System

24 - Overfitting and underfitting
25 - The brute force solution - Grid search
26 - Feature selection

7. Using the Estimator in a Real-World Program

27 - Predict values for new data
28 - Retrain the classifier with fresh data

Conclusion

29 - Wrap-up