GoCareerGo Tutorials

Machine Learning Tutorial

Supervised learning, classification, neural networks and core ML algorithms.

Gradient Descent

An iterative optimization algorithm that adjusts model parameters step by step, moving in the direction that most reduces the loss function.

Compute the gradient (slope) of the loss with respect to each parameter
Update each parameter: param -= learning_rate * gradient
Repeat until the loss stops improving meaningfully
Learning rate too high → overshoots/diverges; too low → painfully slow convergence

for epoch in range(epochs):
    predictions = X.dot(weights)
    error = predictions - y
    gradient = X.T.dot(error) / len(X)
    weights -= learning_rate * gradient