week 2

data science

# -*- coding: utf-8 -*-
"""
Created on Thu Apr 21 21:45:14 2016

@author: ben
"""
import csv
import matplotlib.pyplot as plt

my_file = csv.DictReader(open('train.csv', 'rw'))


#http://scikit-learn.org/

def clean_data(datapoint, average):
    ###fill in missingage and convert to floats
    age = datapoint[0]
    pclass = datapoint[1]
    
    cleaned_up = []
    if age == "":
        age = average
    else:
        age = float(age)
    pclass = int(pclass)
    cleaned_up.append(age)
    cleaned_up.append(pclass)
    
    return cleaned_up
    
#def predict(datapoint):
#    age = datapoint[0]
#    pclass = datapoint[1]
#    if age >= 60:
#        return '0'
#    if pclass == 1:
#        return '1'
#    elif pclass == 3:
#        return '0'
#    elif pclass == 2:
#        if age <= 15:
#            return '1'
#        else:
#            return '0'
        
    

information = []
outcomes = []
predictions = []
for line in my_file:
    datapoint = [line['Age'], line['Pclass']]
    information.append(clean_data(datapoint, 29))
#    pred = predict(clean_data(datapoint, 29))
#    predictions.append(pred)
    outcomes.append(int(line['Survived']))


from sklearn import linear_model

regr = linear_model.LinearRegression()

regr.fit(information, outcomes)

def predict(datapoint):
    pred = regr.predict([datapoint])
    if pred > .5:
        return 1
    else:
        return 0
        
my_file = csv.DictReader(open('train.csv', 'rw'))
   
information = []
outcomes = []
predictions = []
for line in my_file:
    datapoint = [line['Age'], line['Pclass']]
    pred = predict(clean_data(datapoint, 29))
    
    predictions.append(pred)
    outcomes.append(int(line['Survived']))

from sklearn.metrics import accuracy_score
print accuracy_score(outcomes, predictions)
#num_correct = 0.
#for x in range(len(outcomes)):
#    if outcomes[x] == predictions[x]:
#        num_correct = num_correct + 1
#
#print (num_correct/len(predictions)) * 100
#ages = []
#classes = []
#for datapoint in information:
#    ages.append(datapoint[0])
#    classes.append(datapoint[1])
#
#colors = []
#for datapoint in outcomes:
#    if datapoint == '0':
#        colors.append('r')
#    elif datapoint == '1':
#        colors.append('g')
#        
#plt.scatter(ages, classes, color=colors)

python

# -*- coding: utf-8 -*-
"""
Spyder Editor
hjkalskjdhasdk askjdhasldjads 
This is a temporary script file.
"""
import numpy  # ashdjkahdkjalsdh asdhkajsd asdhjkl
import random
#definitions = {
#    'building': 'a thing that has windows and doors',
#    'house': 'a building that people live in',
#    'dog': 'a furry animal with 4 legs that barks'
#}


#definitions['cat'] = 'a furry animal with 4 legs that meows'
#definitions['dog'] = 'a VERY VERY furry animal with 4 legs that barks'
#
##print definitions['dog']
#
#my_dict = {
#    1: [1,2,3], # this sets the key 1
#    2: [4,5,6]
#}
#
#
#students = {
#    'Ben': [90, 100, 87],
#    'Sarah': [90, 85, 68, 100],
#    'Pete': [70, 45, 90]
#}
my_list = [0,1,3,6,9, 20]
my_list.append('asdasd')

print my_list
#for key in students.keys():
#    print key
#    average =  numpy.mean(students[key])
#    goodgrade = average > 85
#    print goodgrade
#    if (goodgrade or key[0] == "P"):
#        print "Good job!"
#    else:
#        print "Could use some work"
#    print '**********'
    
# print goodgrade 
# print max(['abc', 'cde', 'Zzz', 'xyz'])


#myval = True
#counter = 0
#while (myval == True and counter < 5):
#    #do this stuff
#    counter += 1
#    r = random.random()
#    print r
#    if r > .9:
#        myval = False
        
#    if myval is STILL false:
#        jump to line 51 and run again
#    else:
#        keep going