In between juggling my capstone project and lessons, I had not updated the last four weeks of my time at DSI. But here is the final review to make up for it!
The final four weeks were spent learning about Bayesian statistics, Hadoop, Spark, forecasting time series, Flask, Recommender Systems with a brief overview on Docker and MongoDB. Each of these topics is heavy enough you can probably find some full-fledged Udemy courses on just that topic alone. I found the lesson on Hadoop and Spark to be woefully under-allocated in terms of lesson time, while the notebook on recommender systems appeared to be too brief. With capstone deadlines coming up every other week, there was not a lot of time to fully understand these topics. I had to move on. Add them to the list of stuff that I need to revisit after DSI.
My Overall Thoughts
Some pointers I have for those thinking of joining an immersive program:
1) You know how the standard saying goes: “time flies”. Let’s face it, 12 weeks is really not a whole lot of time to get familiar with everything data science has to offer. There is also the ‘memory loss’ that happens when you reach week 12 and realise you are fuzzy on that one or two topics that were covered in week 6. Perhaps others smarter than me have been able to retain all the information, but they could very well have some background in the concepts that were covered (read: mathematics).
2) Preparation for bootcamps start as soon as you make the financial commitment to it. If you are a non-coder like me, you really have to catch up. I think I am a fast learner, but even then I found myself being stretched to the limit in terms of digesting the huge amount of content being crammed into the course.
3) Compared to what I was learning on my own for the past year before DSI, the course has definitely sped up my learning curve. BUT bootcamps are only just a starting point. Learning in the rapidly-changing world of data science never stops. Sure, you will gain enormous exposure to what’s what in the industry, but by no means will you be an expert in them.
4) Practice. Now that I am out of a classroom setting, the onus will be on me to keep coding and trying out new projects. Practice makes perfect… nahhhhh, actually there is no perfect, only ‘better’…
5) Community. My course mates have been great! Our instructor was careful enough to ensure that there was no ‘competition culture’ in the class – everyone was very willing to help one another out. I noticed that beyond the seating arrangement, in the later weeks we gravitated towards those with similar capstone topics to share ideas and knowledge.
TLDR: Data science is a long hard grind just to be reasonably competent, but as long as you are willing to put in the effort, I think you should have no problems. You won’t have everything sorted out even after DSI ends, but that should not be a reason to hold back on a potentially life-changing course!
For those who are so-inclined: my Github page is here.