Created
February 20, 2020 22:10
-
-
Save SimonAB/f0385a15e2eb6d185cc0292a5a845a23 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
""" | |
lin_reg_grad_descent(X, y, α, fit_intercept=true, n_iter=2000) | |
This function uses gradient descent algorithm to find the best weights (θ) | |
that minimises the mean squared loss between the predictions that the model | |
generates and the target vector (y). | |
A tuple of 1D vectors representing the weights (θ) | |
and a history of loss at each iteration (𝐉) is returned. | |
""" | |
function lin_reg_grad_descent(X, y, α, fit_intercept=true, n_iter=2000) | |
# Initialize some useful values | |
m = length(y) # number of training examples | |
if fit_intercept | |
# Add a constant of 1s if fit_intercept is specified | |
constant = ones(m, 1) | |
X = hcat(constant, X) | |
else | |
X # Assume user added constants | |
end | |
# Use the number of features to initialise the theta θ vector | |
n = size(X)[2] | |
θ = zeros(n) | |
# Initialise the cost vector based on the number of iterations | |
𝐉 = zeros(n_iter) | |
for iter in range(1, stop=n_iter) | |
pred = X * θ | |
# Calcaluate the cost for each iter | |
𝐉[iter] = mean_squared_cost(X, y, θ) | |
# Update the theta θ at each iter | |
θ = θ - ((α/m) * X') * (pred - y); | |
end | |
return (θ, 𝐉) | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment