That depends on your educational and work background, and how honest you are with yourself.
As a start, I would suggest that you think in terms of months, instead of weeks, when it comes to setting aside time for pre-course preparation.
I took about 5 months to prepare for my course, during which I took online courses with Codecademy and Data Quest, and studied with a tutor.
Five months may sound like an overkill, but if you have no coding background, like me, then 5 months is frankly not a lot of time to cover the key topics from scratch.
By the time my classes started in March 2019, I had already finished a simple, self-assigned classification project on weather patterns in Singapore.
But even then, it was a daily struggle to keep up with the intense pace in class.
It would have been a complete disaster if I hadn’t spent months going over the basics on my own.
Part of the problem stems from the fact that most students expect a traditional classroom experience and are unfamiliar with how bootcamps are run.
So when lessons race ahead at 100mph, everyone’s shellshocked.
Based on what I’ve seen, bootcamps simply aren’t geared towards teaching you the basics.
Instead, they are designed to accelerate what you are already doing on your own time.
Seen in this light, being well-prepared is not about being “kiasu” as we call such behavior in Singapore.
Rather, it is about making sure you get the strongest possible boost in your learning journey.
#2: CURB YOUR (OVER) ENTHUSIASMBelieve it or not, you will encounter classmates who think it is a good idea to delve into Deep Learning on top of the already over-stuffed curriculum.
If you are a genius, go ahead by all means.
But if you are a regular learner like me, don’t succumb to the “fear of missing out”, or FOMO as they call it.
The reason, while obvious, is not always apparent to over enthusiastic students — it is hard enough to learn the basics well.
Data science is multi-disciplinary in nature.
You may be strong in coding or statistics, but you are unlikely to be equally good at unpacking business problems, visualising data for a general readership, or presenting your work clearly to a restless audience.
Focus on developing the range of skills the bootcamp is trying to impart, instead of pining for those that it is not teaching.
That’s what the rest of your time after the course is for.
One of the most effective reality checks against the tide of FOMO is to read a regular diet of job postings for entry-level data science positions.
What you would notice after a week or two is the sharp disconnect between the online buzz for all things related to Deep Learning, and what employers (at least those in Singapore) are looking for in an entry-level hire — SQL, data cleaning, visualisation skills and basic machine learning know-how etc.
It is good to have some conversational-level knowledge about developments in Deep Learning for sure.
But don’t bite off more than you can chew, not unless you plan on not sleeping for 12 weeks.
#3: TIME MANAGEMENTTime, like traffic in some parts of the world, is relative.
The tongue-in-cheek illustration above is meant to poke fun at how pedestrians deal with the chaotic traffic in Vietnam.
But it is also an accurate depiction of the warped perception of time among some students once the bootcamp gets underway.
The bootcamp will zip by faster than you think, and the 12 weeks or so don’t translate to a lot of time for learning heavy topics like natural language processing.
You would also have to complete several major assignments, including a showpiece project at the end of the course.
Yet some will fall into the trap of thinking they still have time to spare when in fact they don’t.
The lack of time becomes a particularly big problem for the showpiece project, which takes far more time to research, plan, and execute than the two weeks typically set aside for it.
It also leaves very little time for Plan B, if your initial idea fails.
At the start of the bootcamp, your instructor will dutifully inform the class that they should come up with an idea or two for their final project, and start researching on how they can get hold of the data needed for the project.
I can’t think of a better way to make your life easier during the bootcamp than to take that advice seriously and act on it.
I knew what I wanted to do for my showpiece project from the start.
A lot of the grunt work, such as the scrapping, cleaning and formatting of hundreds of thousands of tweets, was spread out over the first several weeks of the bootcamp.
By the time the project topic — on state-backed disinformation campaigns on Twitter — was approved, I had the data and project outline ready to go.
Put your project management skills to work during the bootcamp.
It’s all about managing the risks of potential failure and giving yourself enough buffer to deal with unexpected events — which you won’t have if you start work on your final project at Week 10.
#4 BEING JOB READYAt the bootcamp I attended, there were no exams and no final grades to be handed out upon graduation.
Instead, the school’s focus was on helping the students land a job with three months of graduation — which suited most of us just fine.
But you would be surprised by how easy it is to lose sight of that all-important end-goal once the daily grind of lectures/on-hand-practice/projects gets underway.
This comes into sharp focus when your job search kicks into high gear, and you are reviewing your projects (you could be assigned 3 or 4 of them through the course) to see what you can show to potential employers as examples of your skills and knowledge.
You might be shocked by what you find in your Github repo the second time round.
After wracking your brains and slaving away at a project on home or salary price predictions, it’s natural to want to just dump the notebooks on Github and move on.
That would be cathartic indeed, but comes at a cost.
As someone with no formal work experience in the data science industry, these disparate pieces of project work are among the very few things you can hold up in job interviews to demonstrate your employability (aside from your other paper qualifications).
While the final (or capstone) project is billed as the main showpiece of your capabilities, it might not be enough for more demanding employers who want to see more of your work.
My advice: Take the smaller projects seriously, and treat them as something you can use or repackage as “work samples” in future job interviews.
This might involve the building of a simple web app that utilises the machine learning model you’ve built .
Or you could consolidate and improve upon your data visualisation efforts via platforms like Tableau.
No doubt, it is hard to think that far ahead while you are struggling with an avalanche of class materials.
But the longer term pay off is worth it.
I certainly wish I had approached my project work that way right from the start.
At the very least, make sure your repos and Github account are well organised with clear descriptions of what each piece of work is about.
If you tell yourself that you would clean up your messy repo or refactor your code after the bootcamp wraps up, you can be fairly sure that would never happen.
Get it right from the start, and you’ll save yourself a world of grief.
Do I have misgivings about the bootcamp experience?.For sure.
But that’s a story for another day.
As they say, it’s better to pay it forward than to look back in anger.
If you have specific queries, feel free to reach out via the following channels:LinkedIn: https://www.
linkedin.
com/in/chuachinhon/Twitter: @chinhon.