My intention was to get another blog post in before the New Year. Obviously that didn’t happen – I’ve been on the road for most of November and December. Excuses, I know. But instead of talking about a specific topic, tech or otherwise, I’m just going to give an update on what I’ve been doing since I’ve been home. The whole point of this blog is to write about what I’m learning as reinforcement, and hopefully I’ll get to that. But for now to spur me on, and in the style of Jonah Goldberg’s “Various and Sundry”, this is it:
New “Programming” Book
Sebastian Raschka has a well-known book on machine learning that has lately been available for free. So I figured I would grab it and go through as much as I could in the time I have. And because it was free, I thought I would pay for the book by donating to some open source projects. Michael Kennedy has a great podcast episode on funding core infrastructure for even more motivation. So far I’ve only created a perceptron model for classifying 2D datasets but I think I’m really going to enjoy this book. In addition to Data Science from Scratch it’s a must have.
More Machine Learning
I also started doing Coursera’s Machine Learning Specialization. It’s a six course sequence on machine learning by the University of Washington. They call it a Case Study approach where they show you how to build models in the first course, then go into detail for each one in the subsequent courses. You can read more about it here. I’m about half way through the first course and it’s been enjoyable. It’s easier if you are familiar with Python – even though they use the Python package GraphLab and SFrame instead of scikit-learn and/or pandas dataframes. I may do concurrent projects in scikit-learn. Or just continue with Raschka’s book that uses it.
Being Overwhelmed
Yes, as someone who wants to learn development and data science, I am a bit overwhelmed. This is a common feeling among data science neophytes, and people always say it’s because of the nebulous definition of “data science”. The important thing is that I understand what I bring to the table, and build from there. It’s also nice to have things like Twitter and Stack Exchange sites as sources of encouragement, help, and information. There are no unicorns, only balanced teams. See the appended graphics for a selection of skills.
CheckiO and HackerRank
I feel like I should keep up my basic programming abilities since I have intermittent opportunities to exercise them. I was doing Project Euler for a while (until I approached either my Math, Programming, or time upper bound). Lately I’ve been using HackerRank because it’s the platform used by LaunchCode. I’ve also found interesting problems on CheckiO.
By the way, my work on Project Euler, Coursera, CheckiO/HR, et al is all on GitHub.
Hats
It is a well-known fact (to my wife) that I look ridiculous in a hat. Perhaps that’s why I’ve appreciated the Stack Exchange Winter Bash celebration. Okay, my avatar gets to wear hats, but whatever. I have twelve so far:
While earning these hats (they’re a decent substitution for reputation, right?) I’ve decided to contribute to the Stack Overflow documentation for Anaconda, and if another sponsor will join with us, Bokeh. They are both in a dreadfully underwhelming state.
Things I want to do but will not likely have time for in the near future
MicroPython – I really want to play with this interesting Arduino-like interface (Python on bare metal!)
Rust – The future of C/C++? Well, thread safety sounds nice – data races cannot happen.
Lightweight Django – Here’s a book I would love to use to beef up my web [Django] knowledge.
Well, that about wraps it up. Oh, I made a few applets for my Google Home. One to “find my phone”, another to send a tweet (by voice), and another to have the Google Assistant tell my daughter she should brush her teeth, then call my phone and say it again. It’s very convincing, which is nice since her prior conciliatory nature seems like a distant memory. Like my About Me section says, I spend an inordinate amount of time with my daughter.
Data Science skill set graphics (they’re fun, just look):
