Re-implementing my learnings in machine learning with Python & Python libraries, rather than Matlab/Octave.
Licensed as: CC BY-SA 3.0 https://creativecommons.org/licenses/by-sa/3.0/
Takes 2D data, plots it and fits it using Python libraries. A very simple "Hello World!" level re-implementation of the first assignment.
CSV data is brought in which contains two features and one category: x data (square footage of houses and number of bedrooms) and y data representing the price of the house in dollars.
This information is placed into data structures, then displayed in a 3D plot. The gradientDecent function is used to fit the data and the final fitted value for theta is then printed out.
Below is a 2D rendering of the source data. When the code is actually executed, a 3D plot is generated.
CSV data is brought in which contains two features and one category: x data (student's grades on 2 separate tests) and y data representing whether they were admitted to university or not.
Logistic regression is performed on these 2 features, resulting in a linear best fit boundary estimation between the regions of accepted and rejected potential students.
Below is a 2D plot showing all student data, each with a symbol representing their acceptance or rejection. Additionally, the linearly fitted line of descrimination is also rendered. This line is generated based on the previously calculated logistic regression.
CSV data is brought in which contains two features and one category: x data (student's grades on 2 separate tests) and y data representing whether they were admitted to university or not.
Logistic regression is performed on these 2 features, resulting in a polynomial best fit boundary estimation between the regions of accepted and rejected potential students.
Below is a 2D plot showing all student data, each with a symbol representing their acceptance or rejection. Additionally, the polynomially fitted line of descrimination is also rendered. This line is generated based on the previously calculated logistic regression.
