↓Skip to main content

What next, after part 1 of the Fastai 2019 course?

·3 mins

I began part 1 of the Fastai course sometime in Q4 2019. After an inordinate amount of time for something that I could and should have finished in about 3-4 months, I git pushed my last notebook early last week. Nevertheless, I was excited but then Fastai launched a newer, and possibly a more improved, version of their course a couple of days later!

The new course is apparently better because it incorporates some aspects of traditional Machine Learning (as opposed to how the previous one was only about Deep Learning) and is taught with a newer version of the Fastai library that they launched alongside the course.

So, now I need to decide what side project to pick up.

First, a few tenets or reminders:

  • I won’t learn for the sake of learning. Instead, I’ll be more active during the learning process (similar to how I was with the Fastai course) and keep an eye out on producing something. What I produce could be Today I learned notes on my website, re-implementation of Fastai notebooks or some research papers, or prototypes for a new product that I could launch in future.
  • I reiterate that my end goal will be to produce something.
  • I’ll set stricter deadlines this time.

Second, I only have two major alternatives today:

  • Finish the new Fastai course.
    • While only part 1 of the course is out, I can still complete the whole course through the book.
    • My goal will not be to reproduce all the notebooks myself (like I did the last time). I’ll only reproduce those that will help me learn.
    • The book is only ~600 pages long and I’ve already covered some parts of it, so I should be able to go through it quickly.
    • I want to finish what I start, something that I haven’t done a good job on in the past.
  • Prepare for external interviews.
    • This will take substantial time and isn’t something I can (or even try to) do in parallel to learning Deep Learning.
    • One issue here is that I am not certain when I want to leave Amazon.
      • If I were to move to Canada next year, it’d be a lot safer doing that through Amazon than through another company.
      • I want to stick around in Poirot for awhile to see how my design decisions pan out. At least until GA, which is 1.5 months away. That should be sufficient time to finish the Fastai book.

Third, some more thoughts on those alternatives:

  • I could start with interview preparation now and, if I do accept an offer, I could go through Fastai while the immigration process is underway.
  • I can start with Fastai now and finish it before my parental leave starts. I can then spend whatever time I get during the leave on solving Deep Learning problems/projects and preparing for external interviews.

So, I guess Fastai it is.

↑