My journey started with Signal Processing and Communications, fields very much intertwined with Machine Learning. In fact, my first serious signal processing project was the use of K-means clustering on images (aka Vector Quantization) and later using Neural Networks for binary classification of signals from a data storage channel.
At this juncture in my career in Data Science, I feel like having come across an old friend, however now endowed with so many powerful tools and resources. Libraries developed around the R-Studio are amazing. The Python Scikit-Learn package is quite comprehensive. I really enjoy programming in Python which to me is reminiscent of Matlab (and it's free!). What I really hooked up with is the Ipython (now called Jupyter) notebook. Being able to write notes, produce some code and see the results right beneath in a report format lead to fun AND structured learning. There are so many resources out there for anybody to pick up the necessary skills to be a good data scientist.
I am truly impressed with the ecosystem built around this new exciting field. I think the power of the available material and tools is what enabled me to wade into its deep waters in such a short time. There is so much to learn, but the journey is good.
Before diving into "learning" with the "machine" word in front, I have spent the past year trying to build a language learning system in the form of a startup. To me, the process of learning is just fascinating. Teaching what you have learned in fascination is itself fascinating. Things did not work out as I had planned, but I have learned a lot along the way. The journey goes on.
I intend to post self-assigned projects on this web site using, of course, the Jupyter notebook format. Thanks to GitHub and to the wonderful python tool, Pelican, I am able to post my notebooks as web pages. If you want to be part of this journey, follow along!
This site was prepared using Pelican. I have followed the excellent instructions on this site to make this blog on Github Pages.