I’ve been forever obsessing with Artificial Intelligence and Machine Learning. While browsing through the Udemy courses for these topics, I landed upon Machine Learning A-Z™: Hands-On Python & R In Data Science by Kirill Eremenko and Hadelin de Ponteves. With 473.000+ enrollments, Machine Learning A-Z is worth a thorough review.
Our score:
3.5
| PROS | CONS |
Learn with professionals Interactive exercises Comprehensive Q&A |
No maths are explained Bot-generated answers in Q&A Different instructors Course hasn’t been updated |
Overview
We’re experiencing innovation on a daily basis. These changes require a large chunk of data, which human effort alone cannot handle. Instead, engineers all over the world need to come up with automation to take care of such practices.
Next step, Machine Learning.
Machine Learning is a prominent topic in Artificial Intelligence. The primary aim of the field is to allow computers to learn without human intervention or assistance. From self-driving cars to speech recognition, we’re enjoying the benefits of Artificial Intelligence and Machine Learning. No wonder why Data Scientist has become such a hot career.
To succeed in the area, it’s necessary to choose an appropriate coding language. The two most favored ones are Python and R.
Machine Learning A-Z™: Hands-On Python & R In Data Science is a good place to start with both languages. Taken by 473K+ students, this is the second most popular Udemy course.
Let’s see how the course does in the Pros and Cons list below:
To understand this pros and cons list as well as whether you should take the course, read on for the full review of Machine Learning A-Z™: Hands-On Python & R In Data Science. I’ll be discussing about:
- Kirill Eremenko and Hadelin de Ponteves – the instructors of the course
- The knowledge I’ve gained
- What worked well for me and what could be improved
Meet the Instructors
Let’s talk a bit about the instructors of Machine Learning A-Z™: Hands-On Python & R In Data Science before moving on to the details of the review. The two teachers this time are both professional data scientists with years of experience.
Kirill Eremenko
Kirill Eremenko is a data science entrepreneur, instructor, and consultant. He has got over five years of experience in finance, retail, and transport. He’s also a presenter on Big Data at leading Australian universities and industry events. Kirill enjoys forex trading as it gives him independence personally and financially. These real-life experience and his academic background in Physics and Mathematics help him deliver professional step-by-step coaching in the space of Data Science.
Hadelin de Ponteves
Hadelin de Ponteves is another big name in the industry. He is the co-founder and Director of Technology at BlueLife AI. With a Master’s degree in Engineering, he creates courses covering topics such as machine learning, deep learning, and artificial intelligence. His courses combine the two dimensions of analysis and creativity, allowing you to learn all the analytic skills required in Data Science, by applying them on creative ideas.
Course Content
Around 41 hours, Machine Learning A-Z™: Hands-On Python & R In Data Science is the most extensive course on Data Science on Udemy. It’s packed with content and practical exercises that are based on real-life examples.
The course is open for both beginners and advanced learners. However, you need at least high-school mathematics knowledge for this course to make sense.
There are ten main parts in total, leading you from basic to more difficult topics. The instructors also go through some Python basics assuming the learners have no prior knowledge about the language.
1 Data Preprocessing
The hardest part of machine learning algorithms is their huge set of data. The data needs to be preprocessed into the desired format to make later subsequent easier. This part of the course works with importing libraries and datasets, missing data entries, and categorizing the data into test and training sets.
2 Regression
Once the database is ready, it’s time to apply regression models for future predictions. The Regression section compares the models (Simple Linear, Multiple Linear, Polynomial, Support Vector regressions, Decision Trees and Random Forest classifications) and their performances.
3 Classification and Clustering
The Classification section summarizes the pros and cons of each classification scheme to give a better understanding of the use of each model. In the Clustering segment, the instructors discuss the clustering models, K-Means and Hierarchical Clustering and the basic difference between the two.
4 Association Rule Learning
This part deals with establishing relations between entities. One of the most obvious examples are social media or e-commerce recommendation algorithms (Think Facebook and Amazon). The models used for this purpose are Apriori and Eclat.
5 Reinforcement Learning
Reinforcement learning is a trial and error based method to train machines in AI. It rewards the AI for desired results and punishes otherwise. This happens thanks to Upper Confidence Bound and Thompson Sampling models.
6 Natural Language Processing
Natural Language Processing is most used in speech recognition, text-to-speech conversion, and translation. Throughout this part, learners will evaluate the performance of each classification model as an exercise.
7 Deep Learning
The segment covers Artificial Neural Networks for regression and classification, Convolutional Neural Networks for computer vision, and Recurrent Neural Networks for time series analysis.
8 Dimensionality Reduction
There are two types of Dimensionality Reduction visualizations: Feature Selection and Feature Extraction. As Feature Selection is already covered in the Regression section, this part focuses on Feature Extraction methodologies such as Principal Component Analysis, Lunar Discriminant Analysis, Kernel PCA and Quadratic DA.
9 Model Selection & Boosting
This section discusses the techniques for model selection such as k-fold Cross Validation, Parameter Tuning, and Grid Search. The course concludes with a bonus section focusing on one of the powerful and popular machine learning models, XGBoost.
What’s Good About the Course
Machine Learning A-Z™: Hands-On Python & R In Data Science is the second course I’ve taken from Udemy.
The first one is Complete Python Bootcamp: Go from Zero to Hero in Python 3. I’ve had such a good experience with the Bootcamp that my expectation was higher than ever. The expectation was so high that Machine Learning A-Z didn’t quite reach. I’ll talk about the downsides of this Machine Learning course later in the review.
Still, the course has plenty of things that I’ve enjoyed.
Machine Learning A-Z is a great place to interact with professionals.
As mentioned above, Kirill and Hadelin both got years of experience in the field. To be studying with them is such a privilege.
Occasionally, they engage with students via a podcast, which is quite informal in nature. They’ll go through their backgrounds and a general overview of the course. They also talk about their current projects and other on-going courses. This podcast is a good starting point for those who just join the course.
According to the podcast, Hadelin de Ponteves only sleeps for THREE hours a day (!!)
And he’s been doing that for the last three years. I don’t know about you, but I’ve never met someone like that. Imagine sleeping for three hours a night. Must be hell. And he’s doing it voluntarily? I’ll do anything to know more about this man.
Interactive exercises are a plus.
Another intriguing aspect of the course is the exercises. Students can’t just finish and forget about the exercises. We’re expected to post the solution for the exercises in the Q&A section or PM other members to discuss. Basically, we are to initiate a conversation channel where each solution will not only be evaluated but also discussed upon.
I’m in love with this feature. Online courses are usually taken for granted as there’s no one there to supervise you. You can easily doze off and nobody has to notice.
With these interactive exercises, I have motivation. I know I need to finish and post the solutions to receive feedbacks either from the instructors or fellow students.
Q&A Section is quite comprehensive.
My most valuable take from this course is the Q&A section. It addresses most of the commonly encountered issues. I was able to resolve an issue I faced during installation following the steps mentioned under the Q&A.
Another problem occurred when I tried to install XGBoost on my Windows laptop. But I nailed it using this guide.
Also, the scripts created throughout the curse are written as templates, which can be used in different projects later.
What’s Not Quite Good
As I already did and want to stress more on this review, Machine Learning A-Z™: Hands-On Python & R In Data Science didn’t quite reach my expectations.
For one thing, no maths are explained.
As the course expect learners to have a high-school knowledge of mathematics, it doesn’t provide much exposure to the maths behind the algorithms. However, without understanding the maths, it’s hard to get the real essence of the algorithms. This is more to blamed upon Machine Learning as a field rather than the course as machine learning concepts tend to be technical.
To solve this, the course can either expect learners to refresh the required concepts in advance or the initial sections can be further divided into subsections to make it easier for learners to grasp the concepts.
Though the Q&A section is comprehensive, lots of responses are bot-generated.
The Q&A is great among students. However, the instructors and their team are not of much help.
Oh, didn’t I tell you?
Kirill and Hadelin have a whole SuperDataScience Support team, whose job is to answer whatever questions students may have. But guess what? They don’t.
Most of the responses from SuperDataScience Support are simply “Thank you for taking our course…” They provide no real answer to solve the problems.
In addition, you need to go to their personal website every single section to download the files. I guess this is so that they can advertise for their other courses but it’s such a nuisance for me.
I couldn’t keep track of… the instructors.
I paid for this course expecting to study with Kirill and Hadelin. I did see them, for a short couple of sections. The rest of the course is taken care of by some other teaching assistants, who have a tendency to read from the slides.
In Polynomial regression, one instructor calls it a linear model, while in the next video, another instructor calls it a non-linear model. People are completely confused there and leave lots of comments for their confusion. Unfortunately, the instructors didn’t try to redo the video or make a note to clear that confusion.
This leads to another disappointment:
The course hasn’t been updated.
I used to count on the updating date on Udemy. I shouldn’t have.
Machine Learning A-Z™: Hands-On Python & R In Data Science hasn’t been updated for a while.
The Polynomial regression section is one example. Another is the Decision Tree Regression section.
In this segment, a student mentioned that the technique used is wrong and it should be for Decision Tree Classification and not in Decision Tree Regression. That video is not redone for the last 6+ months from what I can see.
Should You Take the Course?
Machine Learning A-Z™: Hands-On Python & R In Data Science is a good course for anyone who is interested in machine learning or would like to start a career in the field of data science. It’s a good course with some outdated information (especially with the Python tutorials).
If you want to join the course, you’d better revise your math skills beforehand. One thing to keep in mind, though, is that the course will only show you the directions. It’s up to you to make things happen.
The original price of the course is $199. I can ensure you that I’d never take the course at this price. But if the course is discounted to around ten bucks, it’s still worth a try.
Hope that my review about the course Machine Learning A-Z™: Hands-On Python & R In Data Science by Kirill Eremenko and Hadelin de Ponteveson Udemy has been some help.