Since I’ll be spending time in China longer than most of my expat friends around here (until March rather than leaving for Christmas), I’ve been planning to spend my spare time then self-studying. Today’s the day I got started!

A month or so back when I decided to spend this “idle” time in Nanjing instead of at home finding a job, I created a plan for my free time usage in a text file in my notes directory. This will hopefully prevent me from the usual youtube binges. Topics include:

• Machine learning! That deep learning book website, The Elements of Statistical Learning, and a slew of math refreshers.
• Devops/Sysadmin things, like working through the k8 documentation and properly set up system configuration for my servers
• Chinese Language: Keeping up with a TV show, the usual socialization with students, daily news, flash cards, and studying for my HSK on 2019/01/24.
• Chess, or a practice in increasing my time on lichess and decreasing my time in a web browser reading articles. My current metric is to have at least one correspondence game going on in lichess.
• Some personal projects on gitlab like chantrending
• Utilize Emacs – My usual vim usage of directories of notes is getting pretty wacky, so it’s time to grow up and figure out org-mode.

Between all this is the odd job to make money for back home, cooking, and time spent with friends.

My original plan was to start this in 2019 once my friends all went back home, but after waking up at 2pm today after a midnight-3am AoE2 binge (thanks T90official…) and a day of no classes, I tromped over to the local coffee shop for a cheap coffee to get in enough of a manic state to get to work.

Adventures in ML

So today, I finally started on the machine learning topic. Like any respectable computer-science adjacent Millenial Zoomer, I’m starting with Siraj’s “Learn Machine Learning in 3 Months” plan plus some “Super Harsh Guide to Machine Learning” found on reddit. I’m not really looking for a job in ML in the future, but interacting with machine learning algorithms seems like a common theme of future jobs so it’s worthwhile to learn some now while I have the time. Plus I really want to tinker with image recognition software when I finally get my own place.

Since I’m already familiar with calculus, I began looking into linear algebra since I never learned about matrices in college (!!!). I grinded through the first half of 3Blue1Brown’s “Essence of Linear Algebra” videos with a notebook in hand, working through matrix multiplication and determinates. His method of explaining how to think about problems, operations, and transformations intuitively and visually jives much better with my brain than what I’ve worked on before. But I won’t know if this sort of insight is real or illusionary until I need it to solve problems!

After this I looked into Darknet which is a “open source neural network framework written in C and CUDA” written by a very eccentric developer. I’ll look into the source code tomorrow, but implementation will be difficult since what took 0.03 seconds in the example took 63 seconds on my laptop.

After a good start (but not really a solid full day) of study, it’s time to go play pool with some friends at a nearby pool hall. I’ve prepped by watching some youtube videos so I’m feeling pretty good.