sboisen / mlfoundations

supplemental materials for Coursera course on Machine Learning Foundations

Machine Learning Foundations

This repository contains supplemental materials for Machine Learning Foundations from Coursera.

The course is organized to use the GraphLab library. The instructors advocate this for ease of installation and scalability.

GraphLab Create is a highly scalable machine learning library for Python, which also includes the SFrame, a highly-scalable library for data manipulation. A huge advantage of SFrame over Pandas is that with SFrame, you are not limited to datasets that fit in memory, which allows you to deal with large datasets, even on a laptop.

However, after installing GraphLab per their instructions (using Anaconda), i couldn't get their Jupyter notebook to run. So i started translating their notebooks that use SFrames into others using Python Data Analysis Library — pandas: Python Data Analysis Library just so i could complete the assignment. These notebooks should operate correctly using Anaconda 4.3 as the backend.