How to Create Your Personal Data Science Learning Curriculum



Original Source Here

5. Paid Learning Resources on Data Science

Of the paid learning resources that are out there, we’re going to cross out universities and bootcamps as technically they take an instructor and mentor approach and is usually delivered on-site in real-time. The great thing about this approach is the structured curriculum that they provide. Perhaps a topic for another article.

5.1. Books

Books are also a great resource for learning data science by yourself but based on my own learning journey, they might not be the most engaging route. Sure, there are practically a book for almost any topic that you might encounter in data science but I often found it is best used as a complementary learning resource to along with a course (will be discussed shortly) that you’re taking and learning about. I often found books helpful for providing a deep dive into specific topics of my interest. Thus, before we can figure out which book we want to read about, we will first have to first figure out what topics that we want to explore. So to maximize the benefits of learning from books, we would need to craft our own personal learning curriculum.

5.2. Learning platform

Learning platforms are also a great way to learn as a self-taught as they provide structured curriculum that is very much like those provided by universities and bootcamps. Another great thing is the asynchronous approach to learning, which accomodates the life style of working professionals where learners can study and make progress in their own spare time.

Indeed there are several platforms out there and from these which would one should you choose? Frankly, all platforms do provide great content and perhaps the deal breaker lies in the granular differences of each platform. Let’s take a high-level look at the different categories of these platforms that may help you to decide which best suits your needs or learning style.

5.2.1. Pricing

Why pay for courses?
Although it would be nice if courses are free but like all endeavors in life, the creation of a course takes a lot of planning and time to create the content which also entails the expense of hosting video contents on learning platforms.

The creation of a high-quality course takes time and money then how are these financed? On free platforms such as YouTube, part of the advertising revenue goes to content creators as well as from affiliate links (such as the ones in this article), which goes to help finance content creators so that they can create more awesome contents.

As these learning platforms do not serve advertisements then the way for financing the platform that eventually goes to support instructors are the revenue coming from course and subscription sales.

What pricing plans are available?
Learning platforms typically charge on a per course basis (e.g. Udemy) or also on a subscription basis (e.g. Coursera, DataCamp, DataQuest, O’Reilly Learning, Udemy Business, etc.).

Subscription plan — As you can see, most platforms charge on a subscription basis while providing some free course content or free trial period to test out and see if they are right for you. The Pros of this works very much like a buffet, you get unlimited access to courses while the Cons is the steep price that you may incur which is okay if you’re learning new things but often times I find myself paying for a subscription that I may rarely access at times especially during the busy work schedule.

Per course payment — Udemy seems to be the only platform to provide the per course payment model. This seems to be the most flexible as it works in a straight forward manner. You pay for only the courses that you take, no more no less. And for the courses that you paid for, you’ll own access to the course indefinitely. For instance, I will always be able to access courses purchased from a year or two ago. A downside of this may be that you may be accumulating and stashing courses and rarely access them. An easy remedy to this would be self-discipline, consistency and persistence in advancing your learning journey so that you’ll achieve the desired goal that you have set for yourself whether it be to become a data scientist or to upskill.

5.2.2. Instructors

Established and reputable learning platforms such as Udemy, DataCamp, Coursera, etc. are keen of attracting top and talented instructors who are experts in the field and that is well reflected in contents taught on the platform.

For instance, my good friend Ken Jee (who works full-time as a Head of Data Science and part-time as a content creator on YouTube) has created a course on Udemy. Other reputable data science YouTubers and talented educators such as Tina Huang, Daniel Bourke and Giles McMullen-Klein have also created courses in data science on the Udemy platform.

Outside of data science, other exemplar instructors include Angela Yu and Colt Steele who teaches how to code in their widely popular course on Udemy.

Inspired by these talented YouTuber/course creator, hopefully I can find the time and motivation to create a data science course of my own. Let me know in the comments, which course should I create.

5.2.3. Certificates

Degrees and certificates have always been popular amongst working professionals as they are often used as proof of knowledge/skills that they can demonstrate to employers. Several of these learning platforms award learners with a certificate upon completion of a course (e.g. Udemy, DataCamp, etc.) while others upon completion of a series of courses (e.g. Coursera).

As a YouTuber myself (check out my YouTube channel, Data Professor), I always get asked by viewers about the value of certificates. My honest answer is that the knowledge contained in these courses are always beneficial to one’s learning journey. But in order to internalize newfound knowledge and in order to convert that knowledge into actionable skills that inherently become a part of you, then the learning does not stop upon receiving the certificate. Instead, I would always recommend to put the newfound knowledge/skills to practical use by applying them to projects.

6. A Look at the Udemy Platform

Here’s my honest look at the Udemy platform as a paid learning resource for self-taught data scientists.

Screenshot of a course on Udemy.

Let me summarize this as concise bullet points.

  • Price and Flexibility — Firstly, Udemy has 2 versions: (1) Normal plans for typical users that sells on a per course basis (which we will cover in this section) and (2) Udemy Business (which is a organization-based subscription providing. For the normal plan, users can purchase courses and have lifetime access which may be economical when compared to a subscription-based plan. Subscriptions may also be economical given that we take many courses. But if we take courses sparingly then purchasing individual courses may be more cost-effective. It should also be noted that once subcriptions end then access to those courses are lost. Occasionally, some courses may also be offered entirely for free upon entering a special access code
  • Instructors — As mentioned above, course topics spanning data science are taught by talented educators such as Ken Jee, Daniel Bourke, Angela Yu and Colt Steele.
  • Course Topics — Udemy has a large collection of courses spanning not only data science and machine learning but also arts, communication, business, etc. which may help to provide upskilling opportunities to learn about the various technical and soft skills.
  • Course Content — In addition to videos, courses may also come equipped with supplementary materials to aid the learning experience. For example, the Complete Machine Learning & Data Science Bootcamp 2021 by Daniel Bourke also comes with code, workbooks and templates (Jupyter Notebooks) on Github.
  • Instruction Mode — Courses are taught by bite-sized videos that are structured into sections, which may be helpful to jump around to specific topics of your interest. This may very much be similar to that of YouTube but with the added benefit that upon completion of the course, a certificate of completion is conferred. Another benefit is the comprehensive coverage of various topics within the confinement of the course (which may or may not be the case with YouTube videos which may be a standalone video on a specific topic, which may leave us hanging if we want to explore the topic in more depth).

7. Curated list of courses

Here is a curated list of top courses for learning data science categorized according to topics.

Python

SQL

Machine Learning

Model Deployment

Career

8. Translating that newfound knowledge to practical use

Over time, it is certainly great to accumulate knowledge by completing more and more certificates but it is also equally important to put those knowledge/skills to build an ever increasing portfolio of projects. Thus, I highly recommend to build a portfolio website (and GitHub profile) that you can use to showcase your amazing work.

Trust me, this is a game-changer as they not only can serve as a useful resources for other like-minded peers but your work may also inspire those who are embarking on a similar journey. Such joy and excitement that comes from knowing that someone out there is benefiting so much from your projects and contents can help motivate you to keep going in your project sharing endeavor.

In fact, the process of repetitively using that newfound knowledge also helps to reinforce it such that over time you’ll eventually develop mastery. Like the saying goes, practice makes perfect!

9. Time management

In spite of this, I always aim for completion rather than perfection. In my opinion, perfection is our own personal opinion about the work that may or may not be true. For example, what seems perfect to us today may be mediocre to us in 5 years.

I always like to think of it this way. Let’s say that we have a week to complete as task. It may take 20% of that time to reach 80% from perfection while the remaining 80% of the time to perfecting the task. Back in the days, I would aim for perfection and spend that extra 80% of the time making trivial adjustments here and there.

Of course, occasionally you would want to spend that extra time to create a masterpiece but not just all tasks would have to be a masterpiece.

Why?

Because based on my own experience, this may lead to procrastination and possibly burnout. As time is finite, we have to make the most and best use of our time. There’s simply so much to learn and do, therefore we can’t get bogged down or get stuck on one particular topic (know just enough to complete the task and move on).

Conclusion

In summary, we have explored some of the learning resources that one can take to learn data science as a self-taught. Particularly, we have taken a look at some of the free and paid resources and the unique strengths and weaknesses of each while also sharing some of my experiences and tips on how to optimally leverage these resources in learning as well as implementing data science.

Disclosure

  • There may be affiliate links in this article that I may earn from qualifying purchases, which goes into helping the creation of future contents.

Read These Next

AI/ML

Trending AI/ML Article Identified & Digested via Granola by Ramsey Elbasheer; a Machine-Driven RSS Bot

%d bloggers like this: