Coursera-ML

Synopsis

Assignments submitted as a part of Coursera machine learning course July 2014.

Majority of code is written in Octave 3.8.2 and some in 3.8.1.

Following are the topics covered-

1. Linear Regression

a. Univariate linear regression

b. Multivariate linear regression

c. Gradient Descent

d. Normal Equation

2. Logistic Regression

a. Sigmoid Function

b. Regularized logistic regression

c. Vectorized cost function

3. Neural Networks

a. Vectorized logistic regression and gradient descent

b. One-vs-All Prediction

c. Feed forward propagation

d. Backpropagation

e. Regularized neural networks

4. Regularized Linear Regression Bias vs Variance tradeoff

a. Learning curves

b. Polynomial regression

5. Support Vector Machine

a. Gaussian kernel

b. Email Classification

6. Kmeans Clustering and Principal Component Analysis

a. Loyds algorithm of K means

b. Image Compression using Kmeans

c. PCA implementation using SVD

d. Reconstructing approximate representation of data using reduced dimension

7. Anamoly detection and Recommender System

a. Selecting threshold for Gaussian Distribution

b. Preecision and Recall

c. Movie Rating System- COllaborative filtering

Installation

1. Download this repository by

git clone https://github.com/hrushikesh-dhumal/Coursera_Machine_Learning.git

2. Download and install Octave

Example

In each of the exercise folder there is a pdf file with problem description and it tells which file to execute. For Linear regression execute the ex1 from Octave.

Stanford Honor Code

"We strongly encourage students to form study groups, and discuss the lecture videos (including in-video questions). We also encourage you to get together with friends to watch the videos together as a group. However, the answers that you submit for the review questions should be your own work. For the programming exercises, you are welcome to discuss them with other students, discuss specific algorithms, properties of algorithms, etc.; we ask only that you not look at any source code written by a different student, nor show your solution code to other students."

Contributors

Hrushikesh Dhumal (hrushikesh.dhumal@gmail.com)

License

MIT