Online Course popular
The Machine Learning Process A-Z

Name: The Machine Learning Process A-Z Course
Price: 36 USD

with Ken Jee and Jeff Li

Master the complete machine learning lifecycle: from problem definition to model deployment in production

4.8

862 reviews on

Start for free

18,002 students already enrolled

Skill level:

Advanced

Duration:

6 hours

Lessons (6 hours)

CPE credits:

7.5

CPE stands for Continuing Professional Education and represents the mandatory credits a wide range of professionals must earn to maintain their licenses and stay current with regulations and best practices. One CPE credit typically equals 50 minutes of learning. For more details, visit NASBA's official website: www.nasbaregistry.org

Accredited

certificate

What you learn

Acquire real-world skills to deliver machine learning results.
Master all stages of the machine learning lifecycle.
Gain hands-on experience in data preprocessing for ML.
Improve your ML model’s results with advanced feature engineering.
Manage and execute complete ML projects independently.

Topics & tools

Machine LearningModel EvaluationData PreprocessingMachine Learning ProcessData ModelingDealing with Imbalanced DataCross ValidationFeature EngineeringExploratory Data AnalysisMachine and Deep LearningPython

Your instructor

Ken Jee

Data Science Leader | Podcast Host | Z by HP Global Data Science Ambassador

Track record

Bringing real-world expertise from leading global companies

Academic background

Master's degree, Computer Science

Media and recognition

Founder of #66DaysOfData movement
Z by HP Global Data Science Ambassador
Podcast host: Ken's Nearest Neighbors & The Exponential Athlete

Ask Ken a question

Jeff Li

Senior Data Scientist at Netflix

Track record

Bringing real-world expertise from leading global companies

Academic background

Bachelor's degree, Data Science

Media and recognition

Built Netflix's first ad-supply forecasts; drove $1B+ revenue
Led Spotify's podcast ads forecasting; impacted $100M revenue
Co-authored Spotify & DoorDash posts on experiments, ML
Developed Dataquest courses used by 10,000+ learners

Ask Jeff a question

Course overview

Accredited certificates

Course OVERVIEW

Description

CPE Credits: 7.5 Field of Study: Information Technology
Delivery Method: QAS Self Study

Data science education focuses too much on the algorithm itself. In reality, we can have only four lines of code and use them for a variety of problems. The heavy lift of an ML model is the end-to-end process. Jeff Li and Ken Jee walk you step-by-step through this process, so you can successfully take your next project from start to finish. You will learn everything you need to know to set your projects up for success. The Machine Learning Process A-Z course gives you a deep understanding of what machine learning really is. It helps you understand when you should and shouldn’t use this powerful tool. Jeff and Ken break down the specifics of the different problems you can encounter and how machine learning is used in specific domains. In the second part of the course, you will learn the entire modeling process. Jeff and Ken show you how to pull real results and make the ML model work for others, not just yourself. You will learn how to perform essential steps like data preprocessing. In this section, they also show you how to deal with null values and outliers. Next, you’ll see how to explore your data to frame your analysis. Additionally, the course deals with some of the visualization techniques that can help you to see the relationships in your data. After that, we go into feature engineering—one of the most important steps for improving your model’s results. That leads to cross-validation and how to handle bias and variance trade-off in your analysis. Finally, the instructors touch briefly on the model tuning process and how to productionize your work and documentation.

Prerequisites

Basic understanding of machine learning concepts.

Advanced preparation

Curriculum

145 lessons 1 exam

1. Course Introduction

21 min

Machine learning has multiple benefits. The first is consistency—аn algorithm can evaluate hundreds of variables, find indiscernible relationships, and make the same prediction every time. Humans, on the other hand, can become overwhelmed by the amount of information and make erratic predictions. When dealing with high volumes of data, a human may come to different conclusions about predicted outcomes from the same dataset without realizing it.

21 min

Machine learning has multiple benefits. The first is consistency—аn algorithm can evaluate hundreds of variables, find indiscernible relationships, and make the same prediction every time. Humans, on the other hand, can become overwhelmed by the amount of information and make erratic predictions. When dealing with high volumes of data, a human may come to different conclusions about predicted outcomes from the same dataset without realizing it.

Introduction Free

ML Process Course - GitHub repository Free

Meet your instructors Free

How to use this course Free

Additional resources Free

Environment setup Free

Setting up Colab notebooks Free

Setting up notebooks locally Free

Setting up your flashcards - Explainer video Free

Setting up your flashcards - Link Free

Why Machine Learning Free

Why learn the ML process? Free
2. Intro to Machine Learning

15 min

Learn when you should and shouldn’t use machine learning. We teach you when in your organization’s lifecycle to implement ML and what real-life problems you can solve. We also cover the types of ML problems and show you the difference between product analytics vs. data products.

15 min

Learn when you should and shouldn’t use machine learning. We teach you when in your organization’s lifecycle to implement ML and what real-life problems you can solve. We also cover the types of ML problems and show you the difference between product analytics vs. data products.

Real-world ML examples Free

Applying ML Free

When is ML needed? Free

When do we use ML? Free

When to avoid using ML? Free

Real-life ML process Free

Supervised learning vs. Unsupervised learning Free

Regression vs. Classification Free

ML for Product analytics Free

ML for Data products Free
3. The Modeling Process

8 min

The most important step in machine learning is not any fancy technique—it’s properly defining the problem. An incorrect definition can result in months—or even years—of wasted resources. That is why, before we can leverage the power of ML, we need a reason to use it. Whether it’s a directive from a manager, a pain point from a stakeholder, or a gap in the business, you need to properly frame the problem first and foremost.

8 min

The most important step in machine learning is not any fancy technique—it’s properly defining the problem. An incorrect definition can result in months—or even years—of wasted resources. That is why, before we can leverage the power of ML, we need a reason to use it. Whether it’s a directive from a manager, a pain point from a stakeholder, or a gap in the business, you need to properly frame the problem first and foremost.

Problem framing

Understanding the problem

Defining the problem

Business impact

Mapping out solutions

Understanding the data

Checklist
4. Data Collection

8 min

Most projects won’t give you the data in a neat and tidy .csv file like they do in class or on Kaggle. If you work at a big company, it is likely that you will have plenty of data readily available to you. However, you’ll still need to pull it yourself. In this section, we will talk about data collection techniques from most structured to least structured.

8 min

Most projects won’t give you the data in a neat and tidy .csv file like they do in class or on Kaggle. If you work at a big company, it is likely that you will have plenty of data readily available to you. However, you’ll still need to pull it yourself. In this section, we will talk about data collection techniques from most structured to least structured.

Data collection overview

Databases

APIs

Online data capture

Web scraping

Survey
5. Data Preprocessing

35 min

After defining our problem, we need to preprocess our data. Keep in mind that different models have different requirements. For example, linear regression cannot handle null values, whereas some tree-based models can. Additionally, deviant data points can significantly skew the results of some models and not others, e.g., linear models are affected by outliers whereas tree-based models—not as much. In that case, you may want to remove or adjust outliers in order to get the best performance from our models. After all, we want to make sure our data is of the highest quality before we start building our models.

35 min

After defining our problem, we need to preprocess our data. Keep in mind that different models have different requirements. For example, linear regression cannot handle null values, whereas some tree-based models can. Additionally, deviant data points can significantly skew the results of some models and not others, e.g., linear models are affected by outliers whereas tree-based models—not as much. In that case, you may want to remove or adjust outliers in order to get the best performance from our models. After all, we want to make sure our data is of the highest quality before we start building our models.

Intro

Dealing with null values

Types of null values

Missing values - Coding examples

Missing values - Coding portion

Strategies for handling null values

Dealing with outliers

Outliers - Coding portion

Implications of outliers

Detecting outliers

Causes of outliers

Treating outliers

Outliers - Coding examples
6. Exploratory Data Analysis for ML

52 min

Exploratory data analysis is an important prerequisite to the modeling process. With EDA, we get a feel for the assumptions so we can choose the best model and get the most accurate results. The reason we perform this step is to understand the shape of our data and the relationships between the variables.

52 min

Exploratory data analysis is an important prerequisite to the modeling process. With EDA, we get a feel for the assumptions so we can choose the best model and get the most accurate results. The reason we perform this step is to understand the shape of our data and the relationships between the variables.

Overview

Why you need EDA

How to approach EDA

Distributions and single variable plots

Relationships and multi variable plots

Scatterplots

Correlation

Correlation matrix

Bar charts

Line charts

Pivot table

How to generate useful insights

Basic EDA - Coding examples

Basic EDA - Coding portion (1 of 2)

Basic EDA - Coding portion (2 of 2)
7. Feature Engineering

44 min

The feature engineering step will likely reap the highest accuracy gains in our modeling process—even more than the selection of the algorithm itself. The core principle is to find variables that have predictive power. So, regardless of how complex the feature is or how fancy of a technique, if it does not add predictive power to our model, then it doesn’t do anything for your model. Remember: complexity doesn’t create value. Creative and domain-specific features create value.

44 min

The feature engineering step will likely reap the highest accuracy gains in our modeling process—even more than the selection of the algorithm itself. The core principle is to find variables that have predictive power. So, regardless of how complex the feature is or how fancy of a technique, if it does not add predictive power to our model, then it doesn’t do anything for your model. Remember: complexity doesn’t create value. Creative and domain-specific features create value.

Intro

Categorical features

Categorical features - Coding examples

Categorical features - Coding portion

One-hot encoding

Ordinal encoding

Frequency encoding

Target encoding

Probability ratio encoding

Weight of evidence encoding

Binning

Feature engineering for continuous variables

Scaling

Normalization

MinMax scaling

Z-score normalization

Robust scaler

Types of transformations

Logarithmic transformations

Exponential transformations

Square root transformations

Box-Cox transformations

Arithmetic interactions

Binning

Creative features

Continuous feature scaling - Coding examples

Continuous feature scaling - Coding portion
8. Cross Validation

33 min

Cross-validation allows us to test our model on data that the model has not been trained on. If it performs well on this unseen data, we can be more confident that our predictions will generalize well.

33 min

Cross-validation allows us to test our model on data that the model has not been trained on. If it performs well on this unseen data, we can be more confident that our predictions will generalize well.

Cross validation

Train-test-split

Overfitting and underfitting

Bias/Variance tradeoff

Bias

Variance

K-fold cross validation

Leave one out cross validation

Time series cross validation

Monte carlo cross validation

Cross validation - Coding portion

Cross validation - Coding examples
9. Feature Selection

18 min

Feature selection is the process of reducing the model’s complexity by selecting only the highest signal features in the dataset. Typically, there are three classes of feature selection techniques. Wrapper methods require us to exhaustively search through all features and measure performance. They give us the best accuracy but are the most computationally expensive out of the three. Next, we have filter methods that pick up feature properties through univariate statistics. These tend to be faster and easier to check since they don’t rely on cross-validation. Lastly, the embedded methods are baked into the ML algorithm. We cover these in the companion course, Machine Learning Algorithms.

18 min

Feature selection is the process of reducing the model’s complexity by selecting only the highest signal features in the dataset. Typically, there are three classes of feature selection techniques. Wrapper methods require us to exhaustively search through all features and measure performance. They give us the best accuracy but are the most computationally expensive out of the three. Next, we have filter methods that pick up feature properties through univariate statistics. These tend to be faster and easier to check since they don’t rely on cross-validation. Lastly, the embedded methods are baked into the ML algorithm. We cover these in the companion course, Machine Learning Algorithms.

Feature selection

Feature selection - Coding examples

Feature selection - Coding portion

Wrapper methods

Filter methods

Embedded methods
10. Dealing with Imbalanced Data

26 min

How do we deal with imbalance in our data? This is where oversampling, undersampling, and other techniques become useful. We show you how to perform these to optimize your model’s performance.

26 min

How do we deal with imbalance in our data? This is where oversampling, undersampling, and other techniques become useful. We show you how to perform these to optimize your model’s performance.

Intro

Random undersampling

Random oversampling

Synthetic minority oversampling

Borderline SMOTE

Safe-level SMOTE

Adaptive synthetic oversampling

Dealing with imbalanced data - Coding examples

Dealing with imbalanced data - Coding portion
11. Modeling

47 min

In this section of the course, we’ll focus on ML modeling techniques such as the baseline model, model selection, hyperparameter tuning, and assembling.

47 min

In this section of the course, we’ll focus on ML modeling techniques such as the baseline model, model selection, hyperparameter tuning, and assembling.

Modeling

Creating a model baseline

Model selection

Parameter tuning

Ensemble models

The ML modelling process basics - Coding examples

The ML modelling process basics - Coding portion
12. Model evaluation

53 min

We can’t measure the value of our model without proper evaluation criteria. Earlier in this course, we touched on two types of Supervised ML problems—regression and classification. Since these problems use different target variables, we use different metrics to evaluate the quality of the model.

53 min

We can’t measure the value of our model without proper evaluation criteria. Earlier in this course, we touched on two types of Supervised ML problems—regression and classification. Since these problems use different target variables, we use different metrics to evaluate the quality of the model.

Model evaluation

Classification evaluation

Classification metrics - Coding examples

Precision and Recall

Accuracy, Precision, Recall - Coding portion

ROC curves

ROC AUC

F1-Score

PR-AUC

Classification metrics - Coding portion

Log loss

Regression

Regression metrics - Coding examples

R-squared / Adjusted R-squared

Mean absolute error

Root mean squared error

Regression notebook - Coding portion
13. Productionization

7 min

The goal of our work is to make it useful to other people. How do we do that? Through something called productionization. It is a little broader than simply deploying the model into your product; rather it is the delivery of our model to our end user in whatever form.

7 min

The goal of our work is to make it useful to other people. How do we do that? Through something called productionization. It is a little broader than simply deploying the model into your product; rather it is the delivery of our model to our end user in whatever form.

Saving your models

Types of production outputs

Model maintenance
14. Conclusion

6 min

As the course goes through different steps of the modeling process, you’re probably wondering, “How am I supposed to remember each step?” The good news is that building an ML model typically follows a very similar process each time: (1) Problem Framing/Business Understanding; (2) Data Cleaning/Preparation; (3) Exploratory Data Analysis; (4) Feature Engineering; (5) Cross-Validation; (6) Modeling; (7) Evaluation

6 min

As the course goes through different steps of the modeling process, you’re probably wondering, “How am I supposed to remember each step?” The good news is that building an ML model typically follows a very similar process each time: (1) Problem Framing/Business Understanding; (2) Data Cleaning/Preparation; (3) Exploratory Data Analysis; (4) Feature Engineering; (5) Cross-Validation; (6) Modeling; (7) Evaluation

Conclusion
15. Course exam

40 min

40 min

Course exam

Free lessons

1.1 Introduction

3 min

1.2 ML Process Course - GitHub repository

1 min

1.3 Meet your instructors

1 min

1.4 How to use this course

1 min

1.5 Additional resources

1 min

1.6 Environment setup

1 min

Start for free

Very well designed, well narrated course. Starts from scratch and ends up at the advanced levels. The icing on the cake is the integration of My SQL with Tableau at the end of the course which is a very important aspect of today's Data Analytics field. I have personally saved the last 10 practice exercises to dive in some days later, thus it will enable me to revise the course in the process.

Nilotpal C.

See all reviews

9 in 10

of our graduates landed a new AI & data job

after enrollment

With a background in economics and over five years of experience in project management across private and international organizations—including a European Union project in Sudan—I sought new opportunities after relocating to the UAE. Transitioning into data analytics, I began my journey with the 365 Data Science platform, which provided an outstanding and well-structured learning experience. The platform's practical approach and high-quality courses helped me progress from a beginner to a proficient data analyst.

Rabab A.

Before 365: Project Coordinator | SOFRECO

After 365: Freelance Data Analyst

See all reviews

9 in 10

people walk away career-ready

with practical data and AI skills.

I come from a design background, but I wanted to combine creativity with data. 365 Data Science helped me do that. I started with data literacy, then moved into AI and the data analysis career track. The platform kept me motivated with hands-on projects and a learning experience that just made sense. Today, I work as a quality analyst in an AI company, reviewing the outputs of generative models and unsupervised learning. 365 enabled me to see the bigger picture, even beyond my direct role.