Been a While

22 Nov 2017

What’s New?

As usual, it’s been a long while since I last updated the blog. This time, at least, I can say that a lot has been going on, so I’ve been busy! Took our first cross-country trip with the little one a couple weeks ago. It was a lot of fun, with a lot of one-on-one time to bond with the little guy. It was not without its challenges however - feeding a distracted baby when you’re hungry and tired yourself is not the easiest task. I learned and adapted, though. Naptimes were initiated with lengthy strolls around the hotel. I’m sure a lot of the staff were thinking “Oh, there he goes again” but I didn’t mind. The grounds were big enough to provide enough of a variety of scenery. The flights to and from were not too tough either, and that includes diaper changes at altitude! You could say I “earned my wings!”

Other than that, I’ve been cranking away at the most recent online class, which I just finished. This one was an introduction to using PyTorch for deep learning. PyTorch is the new hotness when it comes to deep learning frameworks. After taking this class, I can totally see what the buzz is all about. After getting to know the interface and terminology, the actions of creating neural network models in PyTorch are really straightforward and intuitive - remarkably so, in fact. I’m sure part of that was due to the Instructor, who did a great job of outlining the key steps and giving insight into what is going on when each step is executed. If you’re interested in deep learning and PyTorch, and have a little knowledge of Python, I highly recommend it. Here’s the link to the class on Udemy: Pytorch Class

Class or Practical?

Now that I’ve completed another online class, I have a decision to make. Do I dive back into an online class, or do I find a real-world / practice project to work? Both are important, and provide satisfaction, but in different ways. I often struggle with the real-world projects because they can be very open-ended and/or loosely defined. I do have a couple project ideas in mind this time around, which is half the battle. I think it’s time to see if I can apply these new skills and develop a full-featured project that I can talk about. More on those projects in a (near) future blog post.


The other thing I talk about on this blog with relative frequency is my ongoing search for a new job / career. In the past couple months there has been a promising development followed by a whole lot of nothin’. I have been chatting with the head of a newly formed Data Science group at my company. He as a lot of great ideas about how to use data for business decisions, and many of them line up well with the classes and projects I have been working on in my data science journey. I have helped out with a few coding tasks and even put together a presentation outlining an approach to using deep learning on a satellite imagery recognition task. The sticking point for now, (and foreseeable future) is that there is no line-of-sight for a permanent position in this department. Now that I have a couple years(!) under my belt, I’m confident that if I were able to work full-time on data science projects, I could come up with some really valuable results. Not only that, but I think I’d immediately start feeling some fulfillment at work for the first time in… ever? I can’t get discouraged. I have to keep doing what I’m doing, learning, practicing and putting out excellent work. There will come a day when it all lines up.