Six Point Update

March 10, 2018

Time for another blog post! Quick update on the six main topics I’ve focused on the last 2 weeks:

1. I’ve been consuming data science and python podcasts like no other. The list includes at least:

To all those responsible for making those podcasts - THANK YOU! YOU HAVE MY GRATITUDE!

2. I’ve worked through about 60% of Jake VanderPlas’ Python Data Science Handbook. I am working through the Jupyter Notebook version of the book, which has code samples one can interact with. It covers iPython and four widely-used Python data science libraries: Numpy, Pandas, Matplotlib, and Scikit-Learn. I’ll be honest here. It’s been great at times and a complete slog at others -- the slog part is likely due to my inexperience. I’ve learned many commands that I immediately typed out and put to use during the tutorial, but I fear I will forget them when I begin working on my own projects. I shouldn't be so naive. I'll keep the learnings somewhere in my head, and no one codes without Google handy anyway

I have been applying some of these tools to my own library. Specifically a “gas log” my dad and I maintained in his 1989 Ford F-150 pickup. The log contains information for every gas fill-up since he purchased it brand new - including date, location, number of gallons, and price. Once I get going on the matplotlib chapter (soon!) I’ll have some fun playing with that data set.

3. Began reading Introduction to Statistical Learning by authors Gareth James, Daniela Witten, Trevor Hastie and Rob Tibshirani. I'm wading into Chapter 4 right now. This highly-regarded introductory book to machine learning statistical methods has been pushing my limits of math acumen. Big fan. I aim to apply concepts I learn in this book to modeling projects such as the gas log. The Elements of Statistical Learning by some of the same authors is likely my next statistics book.

4. I updated my script to collect and present the 5 day forecast! Previously I was gathering current weather conditions only.

5. I learned more about Test Driven-Development - aka writing tests before the code itself - but have yet to write tests. I'm sold on pytest as my python library of choice as of right now.

6. Finally, I got dirty learning about .config files. A place to store and access sensitive information like passwords, keys, and usernames. I have begun implementing this with some of my scripts.

And what have I not focused on? I haven’t been committing in Github. I haven’t been working on my own scripts (much). I haven’t put any work into creating my own blog (I did purchase the rights to the domain name!). And finally I have not been collaborating with others (re: open source contributions). All these things feel so important! Logically, finishing both books I'm reading will free up time for the above things, and if I dive into open source work I can use data science tools there.

~ spencer tollefson ~

Leave a comment via a here!