100 Days of Machine Learning

100 Days of Machine Learning

When I started my first developer job last August, I made a promise to myself. Landing the job was the culmination of several years of hard work: learning Japanese, learning how to program, preparing for the FE Exam, and learning how to job-hunt the Japanese way. In my new job I was working in a new environment, in a new industry, and in a second language. I knew it was going to take time to adjust. However, I also knew that once I had got over that initial hurdle it would be all too easy to just coast—to learn just enough to do the job, but nothing more.

My promise was this: after I had settled in to my job, I would continue to learn.

It has now been seven months since I started, and I am as settled as I’ll ever be. So now, it is time to make good on my promise, and I am going to do it in style. From today onwards, I will be doing 100 Days of Code—or in my case, 100 days of machine learning.

The rules are simple:

  1. I will spend an hour a day, for the next 100 days, studying machine learning or writing machine learning code.
  2. I will tweet my progress every day with the #100DaysOfCode hashtag.

I’m going to keep track of my progress on GitHub, using the rayjolt/100-days-of-code repo. And every so often I will write a more in-depth blog post, which will appear on this website.

Mostly I will be doing Andrew Ng’s Neural Networks and Deep Learning course on Coursera, but I may also contribute to open source projects that use machine learning, or do other machine-learning-related coding or study. Today, I will be brushing up on my linear algebra at Khan Academy, because the Deep Learning course doesn’t start until Monday, and because one of the recommended prerequisites is “matrix vector operations and notation”, and it has been a very long time since I learned about vectors and matrices in school.

Wish me luck!

comments powered by Disqus