#allin216: Machine Learning

My wonderful boyfriend, the experienced Network Engineer, loves to tease me about my modern tendencies in online learning. I am his AWS-xpert. (Hard eye roll from me.) He prefers the simplest solution to any problem, but I have the luxury of not being bound by client timelines and can take time to explore any rabbit hole that may prove useful. He and I were both learning Python at the same time, but he applies it to Ansible and I use it for data analysis, increasingly in Jupyter notebooks on a Google Cloud Datalab environment or AWS SageMaker. I’m a bit of a glutton for punishment, so I want to learn enough to be conversationally competent about WAY too many things. For me, it ties in with my desire to work well on teams. Instead of being content with my high school HTML/CSS knowledge circa 2000, I chose to take a Full Stack Web Development course from Colt Steele. And then I added his Advanced Web Developer Bootcamp course out of personal interest generated by the first course, which had already been fairly comprehensive in its own right.

Stepping sideways to the reason I began reviewing web development in the first place, Data Science, and a similar pattern repeats. In pursuit of a School Quality Review position, I took an online course in SAS on the recommendation of my university advisor. I’d learned R for statistics in an undergraduate statistics course taken as part of my math educator license requirements in Utah while working toward my M.Ed. coursework. While I found SAS to be steps below the capabilities I had remembered from R, it felt far superior to the Excel work favored by my employing school district. Learning SAS reminded me how much I love programming. But it also opened my eyes to the existence of Data Science, for which I am most grateful.

The phrase “Data Science” did not exist when I graduated from university with a degree in Mathematics, and statistics was considered by professors in my program to be dry and volatile in comparison with applied mathematics. As a 22-year-old with undiagnosed ADHD and just wanting to enjoy the beauty that my home (the amazing West Coast states) had to offer, I thought I would NEVER want an office job. So I avoided statistics, research, and anything that might include walls and a desk at all costs. Even during the economic recession, I worked mostly hands-on jobs that permitted constant activity: Outdoor Science School Instructor, Lifeguard, coaching various sports teams, and eventually settling into classroom teaching. It was in this most recent/stable career that my ADHD was finally diagnosed and successfully managed. I found myself enjoying the uninterrupted, structured parts of my job more and more.

My brain had been stewing in frustration with the hypocrisy apparent in the K-12 education system status quo to the point that I was ready for a career change. My master’s degree is in Educational Research and Assessment with a certificate in Data-Based Decision Making, so I was initially looking for a research position within the education system. The more I looked into the career change landscape, the more I realized how few research positions were available outside of academia. The SAS course had piqued my interest in considering statistics/CS, so I allowed my search to expand.

Holy moly! What I found outside of academia was overwhelming. Data analyst, data scientist, a whole slew of data engineering positions that could also qualify as software engineering positions…the world was my oyster as a data-head. As a mathematically-adept individual with an affinity for new frameworks, the first difficulty was in identifying a direction. I am an experienced learner, but I know well enough to prioritize what I choose to learn and in what order. To give myself structure in this, I figured a review of my first statistical programming language would be a good start. So looking up courses in R on Udemy is what led me directly to Machine Learning. I chose a Python/R course that seemed reputable and comprehensive. The name? Machine Learning A-Z: Hands-On Python & R in Data Science by Kirill Eremenko and Hadelin de Ponteves.

The first few modules of the course were very familiar and mostly served to add the Python language to my list of new loves. Regression had already been a significant focus of my university coursework in statistics. Classification was the first new concept for me, and these modules opened my eyes to the applicability of data science outside of enterprise. Then Clustering, Association Rule Learning, Reinforcement Learning, Natural Language Processing, and Deep Learning…the further I progressed, the deeper I fell. I was hooked.

Udemy Machine Learning A-Z - UC-ULOBCEJM

By the end of the course, I had branched out into cloud development coursework and chosen a long-term career goal: Data Engineering, preferably finding a niche in which I can use data and development to benefit learners in underprivileged environments. Because I am finally learning that working within the broken education system is not the only way I can try to help improve the situation and opportunities for youth. It is definitely going to be a long road with many wrong turns and dead ends along the way, but I have a growth mindset toward myself and toward my frustrated students. And like my city’s team, I am #allin216.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.