Here's my MATLAB port of that code
data=dlmread('data.csv');
X = data(:,1);
Y = data(:,2);
%Building the model
m = 0;
c = 0;
L = 0.0001; % The learning Rate
epochs = 1000; % The number of iterations to perform gradient descent
n = numel(X); % Number of elements in X
%Performing Gradient Descent
for i=1:epochs
Y_pred = m.*X + c; % The current predicted value of Y
D_m = (-2/n) * sum(X .* (Y - Y_pred)); % Derivative wrt m
D_c = (-2/n) * sum(Y - Y_pred); % Derivative wrt c
m = m - L * D_m; % Update m
c = c - L * D_c; % Update c
end
sprintf('m=%f c=%f', m,c)
This gives the following result
ans =
'm=1.477744 c=0.088937'
You might notice that this is different from the result given in the blog post and that's because there is a bug in the original Python code. The pandas read_csv command that was used interprets the first x,y pair as the row and column names of the data frame. Hence this first data point is not included in the Python analysis.
