robotoil/linear_regression

linear regression

linear_regression_closed_form.py
use linear algebra method to find the best fit model in one shot.

gradient_descent_visualization.py
only one plot is drawn, easy for interactivity.

3plots_1costFunction.py
three plots from different angle plus one cost function plot are drawn.it takes computational power to process.

1, use numpy and sklearn to extract Boston housing prices data. two attributes are chosen .

2, implement gradient descent using python

https://towardsdatascience.com/implementation-of-multi-variate-linear-regression-in-python-using-gradient-descent-optimization-b02f386425b9

3, draw a plane in matplotlib . https://stackoverflow.com/questions/36013063/what-is-the-purpose-of-meshgrid-in-python-numpy the key of drawing a plane is to use meshgrid.

3, use the history of theta(the parameters for this regression model ) and animate 'historical plane' using matplotlib.animation function.

https://brushingupscience.com/2016/06/21/matplotlib-animations-the-easy-way/