class3.py

# -*- coding: utf-8 -*-
"""
Created on Thu Apr 20 20:53:05 2017

@author: ben
"""

# -*- coding: utf-8 -*-
"""
Created on Thu Apr 13 21:51:23 2017
@author: ben
"""
import csv
import numpy

import matplotlib.pyplot as plt

myfile = open('train.csv')
myreader = csv.DictReader(myfile)

avg_age = 29.5

def get_age(line):
    if line['Age'] == '':
        age = avg_age
    else:
        age = float(line['Age'])
    return age
    
def get_feature_vector(datapoint):
    age = get_age(datapoint)
    pclass = int(line['Pclass'])
    if datapoint['Sex'] == 'female':
        return [age, 0, pclass]
    else:
        return [age, 1, pclass]


features_list = []
true_values = []

#survivor_ages  = []
#other_ages = []


for line in myreader:
    
    features = get_feature_vector(line)
    features_list.append(features)
#
    if line['Survived'] == '1':
        survived = True
        true_values.append(True)
    else:
        survived = False
        true_values.append(False)
#
#    if line['Sex'] == 'male':
#        males.append(survived)
#    else:
#        females.append(survived)

#from sklearn.linear_model import LogisticRegression
#mymodel = LogisticRegression()
from sklearn.neighbors import KNeighborsClassifier
mymodel = KNeighborsClassifier(n_neighbors=3, weights='distance')
mymodel.fit(features_list, true_values)

predictions = mymodel.predict(features_list)

correct = 0.0
for i in range(len(predictions)):
    if predictions[i] == true_values[i]:
        correct += 1
        
print 'Accuracy: ', (correct/len(predictions)) * 100