The Machine Learning Process A-Z

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

with Ken Jee and Jeff Li

4.7/5

(547)

There is so much more to data science than just model tuning. This course covers the entire end-to-end machine learning process—from scoping the problem all the way to its productionization.

6 hours 145 lessons

Start course

145 High Quality Lessons

0 Practical Tasks

6 Hours of Content

Certificate of Achievement

Course Overview

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.

Topics covered

Cross ValidationData modelingdata preprocessingDealing with Imbalanced DataExploratory Data AnalysisFeature Engineeringmachine learningMachine Learning ProcessModel Evaluation

What You'll Learn

Many courses jump right into the algorithms, but in this one, Jeff and Ken teach you the real skills needed to leverage machine learning into actual results. You will learn how to perform data preprocessing, engineer your model’s features, and finally productionize your project.

Focus on the end-to-end ML process

Data preprocessing techniques

Data exploration and analysis framing

Feature engineering

Cross-validation

Model tuning and productionization

Curriculum

Course Introduction Free

12 Lessons 19 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
Intro to Machine Learning Free

10 Lessons 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
The Modeling Process

7 Lessons 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
Data Collection

6 Lessons 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
Data Preprocessing

13 Lessons 33 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
Exploratory Data Analysis for ML

15 Lessons 51 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)
Feature Engineering

27 Lessons 42 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
Cross Validation

12 Lessons 32 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
Feature Selection

6 Lessons 17 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
Dealing with Imbalanced Data

9 Lessons 25 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
Modeling

7 Lessons 46 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
Model evaluation

17 Lessons 51 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
Productionization

3 Lessons 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
Conclusion

1 Lesson 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

Student feedback

4.7/5

547 ratings

5 stars

443 (81%)

4 stars

83 (15%)

3 stars

13 (2%)

2 stars

4 (1%)

1 star

4 (1%)

Aidan Au

11.12.2022

This course literarily saves me dozens of hours of time from watching various YouTube videos and looking up codes on the Internet. You’d find this course useful whether you’re doing your first portfolio project, or you’re a working professional. The companion notebooks and GitHub repo are golden! In the past, it took me dozens of hours to google the exact same thing on specific codes. Jeff and Ken did most of the heavy lifting for you, so that you can focus on working on your projects. Most importantly, Jeff has worked in well-known and reputable companies. So you know that his content and material are trustworthy. Though there’re a lot of “Kaggle Champions” on YouTube giving a walk-through on an ML project, I value that Jeff and Ken’s teaching a lot because of his years of experience in major tech companies. This course stands out to me because it talks about the end-to-end process of building ML models. This is a topic that even some bootcamps don’t talk about it a lot. They either don’t spend much time to talk about it thoroughly, or they gloss over it, or they assume that “you would figure it out” along the way. So this course solves that problem to give you a walk through on that process. You would notice that each lecture video’s length is about several minute long to keep it bite-sized. And the coding notebook walkthrough videos are also thorough to cover the lecture video’s portions. So you’ll get both the theoretical knowledge and practical experience in this case. After watching this course, you should be able to do a Machine Learning project from start to finish, end-to-end all by yourself. I look forward to the Part 2 of the course – Machine Learning Algorithms with Jeff Li and Ken Jee.

Mohamed Sherif El-Boraie

30.01.2023

I wanted to provide feedback on the course I recently completed. Overall, I found the course to be good and informative, but I had difficulties completing it due to my lack of understanding of the material. I found that Jeffrey Li's teaching style was not effective for me as he moved quickly through the coding portion without providing sufficient explanation. On the other hand, I appreciated Ken Jee's teaching style as his explanations were clear and helped me gain a full understanding of the material. I believe that Jeffrey may be a better instructor for others, but for my learning style, Ken was the more effective teacher. Thank you for the opportunity to provide feedback and I hope that this information can be used to improve the course for future students.

Paul Figuera

21.05.2023

This course was organized, well laid out and well constructed. Ken and Jeff did a great job explaining some very complicated material and making it uncomplicated. I really appreciated the size of the lectures as it keeps you interested and engaged. They were able to take some complicated functions and explain them in laymen's terms to make them understandable. I am really happy I purchased this course and would highly recommend it to anyone who wants to get into ML. Thank you very much, really appreciate you taking the time and effort in putting together this course.

Jonathan Roman

02.01.2023

The course was informative. As someone who went through a Data Science bootcamp, I still was able to leave this course learning new techniques. The course could use some cleaning up with the order and the notebooks (which I'm not sure was updated since I started, stopped and pick back up the course a few weeks later) but I adjusted accordingly. They appropriate chose the level in which to jump into this course as it is not necessarily for beginners. I'd rate it a 5/5. I look forward to more courses from Jeff and Ken in the future.

lloyd mcleod

28.12.2023

This course is mostly good. The review of each Collab notebook after the videos in each section is very good. One criticism: it seems as if the creators assume the viewer has certain knowledge of some things, thereby never clarifying on some things (for example, using certain terms but never telling its definition). This course of bite-sized videos can be used as a glossary for personal projects you're doing or a companion to other ML courses that dive deeper into certain subjects.

“Companies don’t pay data scientists all this money to just write "model.fit" and "model.predict()". Having a thorough end-to-end process is how data scientists create business value for their employers. This is also how a data scientist levels up to a senior and eventually principal DS.”

Ken Jee

220K Subscribers on Youtube

Courses You May Like

The Machine Learning Process A-Z

with Ken Jee and Jeff Li

Start Course

The Machine Learning Process A-Z

Course Overview

Topics covered

What You'll Learn

Curriculum

Courses You May Like

SQL + Tableau + Python

Deep Learning with TensorFlow 2

Machine Learning in Python

Convolutional Neural Networks with TensorFlow in Python

Linear Algebra and Feature Selection

Machine Learning with Naïve Bayes

Machine Learning in Excel

Machine Learning with Support Vector Machines

Machine Learning with Decision Trees and Random Forests

Machine Learning with K-Nearest Neighbors

Machine Learning with Ridge and Lasso Regression

The Machine Learning Algorithms A-Z

Machine Learning Deep Dive: Business Applications and Coding Walkthroughs

ChatGPT for Data Science