/Machine-Learning-Specialization

My Assignments of Coursera Machine Learning Specialization using Scikit-Learn, Pandas, Numpy and Scipy

Primary LanguageJupyter Notebook

Machine Learning Specialization - University of Washington

The contents of this specialization is quite nice for a noob like me. Last month when I finished Andrew Ng's Machine Learning on coursera, I decided to try CS229. However I found the gap between these two courses is quite large, with the fact that both of these two courses are named as Machine Learning. Literally two different courses.

So I take this specialization as a transition and also to solidify my knowledge.

About this course


I will start with the most disappointed. It suggests you to use GraphLab instead of other common open source libraries like pandas and scikit-learn. In the first course, I used GraphLab and SFrame, but in the other three courses I used pandas, numpy, scipy and scikit-learn. If you want to know about the popular libraries for machine learning, I suggest you also use these libraries. Another con is the depth. For example, Hessian matrix is not introduced in Regression, bagging is not included in ensemble and no derivation of EM in GMM.

But with all the cons above, I will still recommend this course to a beginner. Happy Learning.

Last, huge thanks to Emily and Carlos, especially Andrew.

Environment and Libraries


Windows 10 64 bit - Anaconda 3 - Jupyter

Pandas, Numpy, Scipy, Scikit-Learn

Notes


Last again, u can find my notes on my blog SYCabin The notes may not be comprehensive because they are based on my own former background. I will be happy if it could help.

PS: some of the course material is written in python2 while some in python3.