logistic_regression.py

import random
import math
import csv 

def hypothesis(params, features):
    z = sum(p * f for p, f in zip(params, features))
    return 1 / (1 + math.e ** -z) 


def logistic_regression(learning_rate, samples):
    params = [random.random()] * len(samples[0].features)
    for s in samples:
        error = s.target - hypothesis(params, s.features)
        params = [p + (learning_rate * error * f) for p, f in zip(params, s.features)]
    return params

class Sample(object):

    def __init__(self, target, features):
        self.target = target
        self.features = features

if __name__ == '__main__':
    with open("wine.data") as f:
        record = csv.reader(f)
        rows = (map(float, row) for row in record)
        samples = [Sample(r[0], r[1:]) for r in rows]

    learning_rate = 0.1 
    params = logistic_regression(0.1, samples)
    for i, s in enumerate(samples):
        print i, hypothesis(params, s.features)