Online machine learning course for R users

1 minute read

A while ago I took an online machine learning course offered by Andrew NG on Coursera platform. The course was spectacular although all the assignments MUST be submitted via MATLAB/OCTAVE. This would be a tedious effort for users who have spent a lot of time learning the basics of R language. Thus, I developed all the starter codes for R users in order to be able to carry out their assignments in R and submit them to Coursera directly from R. For more information visit the course Repo on Github.


A few screen-shots of the plots produced in R:

Anomaly Detection Gradient Descent Convergence K-Means Clustering K-Means Raster Compress Learning Curves PCA Face Dataset SVM RBF Kernel Multiple Regression PCA Pixel Dataset Centroids

Topics covered in the course and assignments

  1. Linear regression, cost function and normalization
  2. Gradient descent and advanced optimization
  3. Multiple linear regression and normal equation
  4. Logistic regression, decision boundary and multi-class classification
  5. Over-fitting and Regularization
  6. Neural Network non-linear classification
  7. Model validation, diagnosis and learning curves
  8. System design, prioritizing and error analysis
  9. Support vector machine (SVM), large margin classification and SVM kernels (linear and Gaussian)
  10. K-Means clustering
  11. Principal component analysis (PCA)
  12. Anomaly detection, supervised learning
  13. Recommender systems, Collaborative filtering
  14. Large scale machine learning, stochastic and mini-batch gradient descent, online learning, map reduce