In this post, I would like to share the online course on Data Science that I started my journey with.
It is neither the perfect course nor the end all be all of Data Science, but it is the course I started with and it, therefore, holds a special place in my heart.
Caveat: This blog post contains affiliate links, which means if you buy the course through my link I will earn a small commission which helps me keep this blog running, with no additional cost on the course to you. If you prefer not to, just don’t use my link and go find the course directly on the Udemy website. I sincerely like the course and that’s the reason I suggest it. If you feel better not to use my link there are no hard feelings :).
The course is called Python for Data Science and Machine Learning Bootcamp and is offered on Udemy by an instructor named Jose Portilla.
How was I introduced to the course
When I first applied for the PDEng program I am currently a part of, this course was suggested to me by one of the trainees of the program at the time. And I am really grateful that it was. There are several reasons for that.
Why I suggest Python for Data Science and Machine Learning Bootcamp
To start with, the course is literally a no-excuse one in terms of cost, since you can find it for as low as 12 euros in the sales that Udemy regularly has.
That being said, it doesn’t mean that the course lacks in quality. On the contrary, it is a very detailed course that touches upon most of the basic aspects of Data Science and Machine Learning.
To take the course, it’s good to have at least a basic background in programming and some math/statistics fundamentals. Apart from that, the course has a very low entry barrier.
Jose Portilla is one of my favorite instructors on Udemy and after this course, I have taken more of his courses after this one and he never lets me down. He explains everything very clearly, the quality of the videos and the sound is very high, and his approach is a no-nonsense one.
He does not bother you with more than the necessary and that is something I personally appreciate when I want to get up to date fast with some topic.
Style of Teaching
For this course, he gives small introductory videos on the intuition/theory behind the concepts and then implements everything in Python using the modern library packages of the Python ecosystem.
He goes all the way and codes live on the screen, explaining every little bit of code. He also provides all the code in Jupyter Notebooks on Github and also in the form of zipped files.
More specifically, through this course, you can get acquainted with NumPy, pandas, matpltolib, seaborn, some plotly/cufflinks, sci-kit-learn, and a bit of Spark.
The course doesn’t go very deep in all these topics, but of course, it is logical for an introductory course.
That being said, it goes fairly in-depth in NumPy, pandas, and matplotlib/seaborn, which are at the core of Data Science. With the aforementioned packages, you can do data manipulation and data visualization, two concepts that are at the heart of Data Science.
The Importance of the Non-Fancy in Data Science
At this point, a small note. I see many people around being very excited with Data Science and they basically mean Machine Learning and Deep Learning.
While the two topics are very cool and indeed integral parts of Data Science, I personally would like to share that I have found Data Manipulation and Data Visualization skills equally, if not more, important.
They may not sound that fancy at first sight (even though with Data Visualization especially you can do very fancy things) but they play a key role in almost every Data Science and Machine Learning project. Especially Data Manipulation is the dirty worker you absolutely need to fuel your data-hungry Machine Learning/Deep Learning/AI/whatever algorithms.
So, do not underestimate the importance of these skills. It is also directly connected to the meaning of Data Science that I blogged about in a previous post. But let’s not deviate more.
In terms of knowledge, the course covers several topics and serves as a good introduction to them. Namely:
- Data Manipulation and Preprocessing
- Data Visualization
- Core Machine Learning Algorithms (both Supervised and Unsupervised)
- Touches upon the more fancy concepts, such as Natural Language, Deep Learning, Recommender Systems, and Big Data.
Who is this course for?
I would ideally suggest this course to someone who is first, new to Data Science, and second, has at least a bit of a background in programming and math/statistics.
If you are someone already experienced with Data Science and/or Machine Learning, probably this course won’t have that much to offer.
Unless you want to get into Python and/or feel you lack in some of the above topics, if you are already experienced probably you are not really in a need for it.
On the other side, if you are someone with no background in programming and or math/statistics I would say that starting off on Data Science and Machine Learning, even if it is a beginner-friendly course, is probably out of scope.
Go back and grasp the fundamentals, that is programming principles/computer science and basic linear algebra, calculus, and statistics, and then come back to this course.
Learning Data Science and Machine Learning is an ongoing Journey
With all the above being said, I would also like to share at this point my view on online courses and learning Data Science in general.
Never believe that you will do an online course and you will be fully covered. No online course can really cover everything there is to know about this broad field we now call Data Science.
In order to become good in this field, you should start feeling comfortable with the unknown and nurture the ability to thrive in chaos.
The more you learn in this field the more you will understand how much you don’t know.
For that reason, get into the habit of lifelong learning. Absolutely never stop learning and become complacent.
The moment you become complacent is the moment you stop growing. It’s not a coincidence that most technology companies invest in the personal development of their employees.
They make available a lot of money, as well as time (in my program it is as high as 15 hours of the 40 per week) in personal development that the employee can use in order to grow their skills the way they see fit.
So, don’t expect this course will make you a full-fledged Data Scientist. It will require a lot more time in the trenches of messy real data in order to achieve that.
Invest in Learning Data Science
To conclude, these days the entry barrier for Data Science is very low. You can start learning from top-notch instructors like Jose very cheaply in the comfort of your home.
It’s the best time to capitalize on that, especially when gems like the Python for Data Science and Machine Learning Bootcamp are around.
The course is cheap, the instructor is amazing, the content is generous. No excuse not to take it really. Be sure to click the image below and check it out. 😉
Until next time,