Hi all! It’s been a while. I’ve been working on Kaggle and my thesis. Really hard stuff.
This week we’re starting Deep Dive Wednesdays in our team in Trend Labs. It’s a series of workshops that I will facilitate where we tackle data science topics and machine learning. It will be hands-on, with the scipy stack and scikit-learn as our tools. I’ll upload some of our outputs in the coming weeks.
The course content and my students
In our team, we have regular Techie Tuesdays where we discuss technology trends and techniques that may be helpful for our team. I lead 1 software architecture session and 3 data science sessions in Techie Tuesdays so far. That we had 3 data science sessions is already a sign that the material is important enough to need its own in-depth workshops. Besides, number crunching and AI superintelligences always deliver the headlines.
Seriously though, data science is a skill worth learning.
My students are familiar with only the basics of machine learning, having tackled some statistics, regression and supervised learning in my previous sessions. Now it’s time to go to the next level, hands-dirty, data wrangling time.
I’ll be using some slides of Andrew Ng and Pros Naval, my professor in UP Diliman. I’ll be referencing some nuggets of knowledge I picked up over the years. There will also be sweet homework and romantic case studies. (Happy Valentines day, by the way!)
Now, week 0
I wanted to share with the interested team members my expectations for the course, as well as give them a hand in gauging their interest and commitment for learning. Week 0 gives us this chance. Here’s how I launched the session.
These insights are from Deloitte.
Activity Time
I gave the students an activity. Three groups are tasked to find interesting things in the Tableau dashboards I provided. First is the Facebook/Internet coverage. Second is the gender equality and child mortality indexes. Third is the world hunger index. The following is a quick mashup of how they did.
I picked these topics since firstly, communicating insights from data is key to any data scientist’s career. Secondly, these dashboards convey important stories around the world, particularly UN Millennium Development Goals. The reality that we’re inching closer to ending hunger and gender equality proves that humanity is doing something right, albeit we still have a long way to go. But we’re getting somewhere.
Closing Time
Finally, I closed the session by having a little plug on math. Mathematical expressions are not a lofty past time. It is a way to elegantly explain complex thought in ways that could be implemented in different fields and even in different decades. As Neil Degrasse Tyson said, Albert Einstein didn’t know that his theories lead to… satellites!
I think I hit something there on the part of stereo cameras. Depth estimation is, at its core, triangles. Geometry. Beautiful. I hope that struck something on you camera aficionados as well!
The coming weeks will be heavy. But let’s JFDI!
It is the REAL DYL 