This repository contains some machine learning projects as a practise on machine learning course on Coursera for Prof. Andrew Ng from Stanford University.
- PC
- MATLAB or Octave
- Project1: Purchase Prediction Project :
*This project is an application on Classification using Logistic Regression algorithm. We have dataset for people who purchased a specific production. We have ID, Gender(1 for males and 0 for females), Age, Salary and purchase status (1 for purchasing and 0 for not). we need to classify it by Age and salary for purchased and not purchased people and develop our prediction model depend on the given dataset. I used people whose age is dividabe by 2 as a Training Set and tested my model by people whose age is not dividabe by 2 as a Test set.
- To test this run Purchase.m script on MATLAB or Octave, but make sure that all files in your directory.
- After that you would show 2 figures: Figure1 and Figure2.
- Figure1 Training set with output classifier: shows the result of training set with the classifier which all points above the line most probable to purchase and under the line most probable not to purchase.
- Figure2 Test set with output classifier: shows the result of test set with the classifier which all points above the line most probable to purchase and under the line most probable not to purchase.
- I also calculated the accuracy of the model after testing the training set and print it to the workspace.
- You can find all assignments here.
- Assignments 1: Application on linear regression, cost function, and gradient descent. And this is Project description.
- Assignments 2: Application on logistic regression, cost function, gradient descent, and regularization for solving overfitting problem. And this is Project description.
- Assignments 3: Application on one-vs-all logistic regression and neural networks to recognize hand-written digits. And this is Project description.
- Assignments 4: Application on the backpropagation algorithm for neural networks and apply it to the task of hand-written digit recognition. And this is Project description.