Must Love Data Mondays: Machine Learning Foundations: A Case Study Approach by @UW

Last week, I wrote about a few new coursera courses I discovered but then I found the Machine Learning specialization on Coursera which is offered by the University of Washington.

Last week, I started the specialization and the first week consisted of a quick introduction to Python, because the course is being taught in Python, installing GraphLab Create, and iPython Notebook.

The course recommends that you should be familiar with one other programming language. I completely agree because the tutorial was very brief, used one example, and lasted maybe thirty minutes. The Python tutorial was probably more useful for people who are using iPython Notebooks for the first time because all of these tutorials are done in iPython.

Another thing to take note of is that this course uses GraphLab which is a paid software that students in the specialization are allowed to use for free. A number of students in the discussion rooms have been complaining about it because it’s expensive to acquire a license and that we should be learning to use free software instead.

I understand this agreement and I’d rather use free software but this specialization is sponsored by Dato, which sells GraphLab, so they’re going to want us to use they’re software.

The creators are also trying to make this course as accessible as possible to people who may not be experienced in setting up a VM box with preinstalled tools like the popular UWash Intro to Data Science course and Dato probably helps beginners to get up an running quicker since this is only a six week course.

The use of GraphLab hasn’t deterred me from trying the course. The main reason I enrolled in this course was its focus on machine learning projects.  In the first course of the specialization I’ll be creating projects to predict housing prices, analyze sentiment, retrieving documents based on similarities, recommending products, and searching for images.

I’ll be writing about each project I complete through this course.  Next week, I’ll talk about the predicting housing prices project.