Perceptron

theta = [0, 0, 0, 0, 0] #N-dimensions large vector of zeros

theta_naught

while True:
    changed = False
    for i in range(N):
        if y[i] (theta.transpose() * X[i] + theta_naught):
            theta = theta + y[i] * X[i]
            theta_naught = theta_naught 
            changed = True

    if not changed:
        break

Related

• 21 Perceptron Algorithm and Logistic Regression — same perceptron update rule, extended to logistic regression