Make sure you have data.
That’s right, before you can use your arsenal of tools to inspect, clean, and visualize data, you must collect it. Okay, I’m sure that any decent data scientist knows this. But when you’re a wannabe data scientist like me, it’s a constant reminder.
I’ve been meaning for a while now to start working on machine learning project that uses linear regression to predict the weather. Except I didn’t have any local weather data to feed into a linear regression model. So I did, or at least that’s in progress. I’m getting my weather data via a Wunderground.com Developer API and I’m limited to 500 calls per 24 hours, so I’ll finish that tomorrow. I also need to revise my code, and really, I just did a poor plan of planning. This leads to the first rule of blogging about data science. Plan your blog posts!
Again, a seasoned blogger doesn’t have this problem, but a wannabe needs ongoing supervision. I was about to write a post about my recent Kaggle competition submission, but I realize I need to plan it out a little bit- write an outline, identify some good code snippets, etc. But I didn’t so that is going to have to wait until next week. Fortunately, for you however, you don’t have to wait until next week for a blog post.
Finally, I give you the second rule of blogging about data science (or anything, really): just write! Sometimes, you just have to write, even if it’s not the most planned out masterpiece. If I waited until I had most thought out and well written article worthy of the Pulizter, I would have a blog with zero posts. Which is why you’re reading this nonsense that I threw together after midnight instead of something a bit more serious. Don’t worry though, that will come soon enough. Thanks for reading!
Don’t mind me, I’m just rambling.