In this subject, we learn how to use Python to deploy machine learning algorithms. This course is designed for people with no programming experience. We start with the basics of Python before moving onto courses dedicated to building machine learning algorithms for a variety of business use cases.
The courses in this track focus on getting familiar with Python without stepping into machine learning. Once you've completed these courses, you'll be ready to learn how to build machine learning algorithms in Python.
1. Python Fundamentals
In this course, the learner will get a grasp of the basics of Python. Although the goal of this course is to take the first step in a path towards building machine learning algorithms in Python, we'll learn some important general insights about Python along the way.
2. Functions, Conditionality and Loops
In this course, we learn some important Python functionality that considerably enhances what we can do with Python code. We learn about functions, conditional statements, and repeating code loops.
3. Storing, Transforming and Exploring Data
This course focuses on using Python to organize datasets for use in machine learning algorithms. We’ll learn about ways to load, clean, transform and visualize our data. We’ll also cover 3 critical Python libraries: NumPy for advanced calculations, Pandas for reading and manipulating datasets, and MatPlotLib for visualizing data.
Machine Learning in Python
In this track, we explore how to use Python to build linear regression, decision trees, and K-means clustering algorithms.
1. Linear Regression in Python
In this course, we deploy our first machine learning algorithm. This is a supervised machine learning algorithm meaning that it can be used to predict future values. We’ll discover the logic behind linear regression and then explore how it can be deployed to predict numeric values. We’ll turn our attention to a specific business use case that will become the basis of our first linear regression model.
2. Decision Trees in Python
In this course, we focus on classification. These algorithms are used to predict outcomes that can only have a few possible variations. For example, it can be used by a subscription-based company to predict whether a customer will cancel, stay, downgrade or upgrade. There are many classification algorithms, but we’ll focus on Decision Trees which are both easy to understand and to visualize.
3. Clustering in Python
In this course, we deploy a clustering algorithm. Unlike supervised learning techniques, this can't make future predictions for a specific value, but it can be used to find patterns within our data. We’ll use these to segment customers into separate clusters. This will allow the business to tailor its responses to these customers. There are many clustering algorithms, but we’ll focus on K-Means clustering which is both easy to understand and to visualize.