Skip to content
R code for the assignments of Ng's Coursera machine learning course
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
Instructions(pdf) Initial Commit Apr 15, 2016
Snapshot merging stater and solution folders Jul 12, 2018
Starter_solution merging stater and solution folders Jul 12, 2018
.gitignore merging stater and solution folders Jul 12, 2018 merging stater and solution folders Jul 12, 2018


This is the R version assignments of the online machine learning course (MOOC) on Coursera website by Prof. Andrew Ng.

To download lecture videos visit the course website:

This repository provides the starter code to solve the assignment in R statistical software; the completed assignments are also available beside each exercise file.

Simply follow these steps to complete the assignments:

  1. View the lectures
  2. Read the instructions (pdf). Instructions are basically for MATLAB/OCTAVE. R compatible version of instructions will become available here as Wiki pages in future.
  3. Use the Starter_solution folder and fill the parts of the code that is written "YOUR CODE HERE"
  4. If you couldn't solve it yourself, get help from the accompanied file suffixed by _solution inside the same folder of the starter code. For example, starter_solution/ex1/computeCost.R has an associated solution file named starter_solution/ex1/computeCost_solution.R
  5. Submit


In order to produce similar results and plots to Octave/Matlab, you should install a few packages:

  • rgl package is used to produce the 3D scatter plots and surface plots in the exercises.

  • SnowballC: portStemmer function in this package has the same role of the portStemmer.m.

  • raster package is used to produce the plot of the bird in exercise 7.

  • jsonlite and httr packages are needed for submission.

  • pinv.R: The ginv function, generalized inverse, in MASS package doesn't produce the same result of the Matlab pinv (pseudo-inverse). pinv.R is the modified version of MASS ginv to produce the same effect of the MATLAB pinv. For more info see the stackoverflow discussion

  • lbfgsb3_.R: Certain optimization tasks could only be solved using lbfgsb3 package, yet there are a few bugs in this package. The purpose of lbfgsb3_.R is to address these bugs; it is used for exercises 4 and 8. Beware that fmincg/fminunc optimization functions in Matlab takes one function as input and computes cost and gradient simultaneously. However, cost and gradient functions MUST be supplied into optim or lbfgsb3 functions individually.

Before starting to code, install the following packages: install.packages(c('rgl','lbfgsb3','SnowballC','raster','jsonlite', 'httr'))

Note that you don't have to do anything with what is mentioned above, just be informed.


After completing each assignment, source("submit.r") and then submit() in your R console.

I submitted the solutions to Coursera for testing and the scores were 100%. Please report any problem with submission here.

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, on-line learning, map reduce


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

You can’t perform that action at this time.