× {{alert.msg}} Never ask again
Class time and plan

Live Mentor

Small Group

Hands-On Learning

“I'll show you how to use machine-learning to tackle real-world predictive analyses. We'll start with industrial-strength data sets. We'll break them down and understand them. We'll see what commercially valuable models we can extract.”

Peter Figliozzi, Ph. D.

Pete is a professional data scientist. He has used machine-learning algorithms to create predictive engines in variety of fields including marketing, mechanical prognostics and health management, and algorithmic stock trading. Pete enjoys Codementoring and the great interaction and learning that comes with it. Pete's degree is in Physics from the University of Texas, Austin.

Learn Machine Learning with a Live Mentor, Not with Video Tutorials

Who is this class for?

This class is for anyone who wants to learn how to apply machine-learning to real-world (commercial) problems. All you'll need is a basic, working knowledge of Python. No machine-learning experience is required.

What will you learn?

By the end of the class, you'll be able to: analyze a real-world, industrial-strength data set; explore that data set; formulate a commercially viable predictive question; and apply machine-learning algorithms to create a predictive engine.

Codementor Machine Learning Class: Syllabus

Lesson 1

College Degrees

We'll use the U.S. Department of Education College Scorecard dataset to build a college recommendation service. 

This introductory lesson will cover the essentials: setting up and working with Jupyter notebooks; importing, cleaning, and assembling real-world data with the Pandas library; displaying inline graphs and charts with Pandas and matplotlib. 

We'll use Pandas together with Sci-kit Learn's preprocessing and clustering modules to characterize the data.


Lesson 2

College Degrees Part II

We'll use our findings from Lesson 1 to construct the college recommendation service. 

We'll learn and apply Sci-kit Learn's principal component analysis (PCA), together with a suitable clustering algorithm. 

Once we finish, we'll show how to take code out of the Jupyter notebook, respond to JSON-formatted queries, and provide a response.

Lesson 3

Public Opinion

Using a full month of Reddit comments, we'll use some natural language processing (NLP) techniques to infer public opinion on different types of cars. 

We'll discuss the Python NLP landscape, using the Natural Language Toolkit (nltk) for sentiment analysis, text pre-processing, and descriptive statistics. 

We will construct sparse-vector (bag-of-word) models, apply term-frequency inverse document-frequency TF-IDF weightings, and retrieve similar documents using cosine-similarity. 

We will also construct more advanced models including Latent Dirichlet Allocation (LDA), Latent Semantic Indexing (LSI), and paragraph vectors/doc2vec using the gensim library. 

We will demonstrate LDA visualization and other document visualization methods.

Lesson 4


We'll use data from an online lender to build a prediction engine for default/late-payment risk. 

We will use Sci-kit learn to evaluate our input variables for predictive power and build a random-forest classifier. We'll learn how to visualize our model using graphViz and evaluate its effectiveness using receiver-operator characteristic (ROC) curves.

Bonus: Online advertising-- we'll explore Bayesian statistical models and A/B testing.

You Will Get

8 Hours of Live Classes

The weekly live group lecture is the core of the class. The mentor will teach the curriculum live and students can ask questions directly.

Hands-on Exercises

You'll get the chance to code live via hands-on exercises. All students will work on the exercises together and get instant feedback from the mentor via screen-sharing.

Unlimited Lesson Playbacks

Had to miss a class? Fear not. All lessons are recorded and you can review the recordings in the future.

Frequently Asked Questions

How is this different from other courses?

For only $99, you’ll get to learn Python live with a real online mentor. This includes 8 hours of interactive lectures and hands-on exercises. Having live interaction with a mentor makes learning much easier than watching recorded video tutorials.

What do I need to know before the class?

This class is for anyone who wants to learn how to apply machine-learning to real-world (commercial) problems. All you'll need is a basic, working knowledge of Python. No machine-learning experience is required.

What if I miss a class?

All classes will be recorded and you‘ll have access to the recordings at any time. You can also ask questions in our Slack channel and get help from other members of the community.

What is a Delayed Live Class?

Delayed Live Class is for those who couldn't make it for the live sessions. With Delayed Live Classes, you'll be able to watch recordings uploaded within 24 hours after each session, and have access to our Slack community to ask questions and discuss with your peers and the instructor.