Kaggle:Titanic: Machine Learning from Disaster

一直想着抓取股票的变化，偶然的机会在看股票数据抓取的博客看到了kaggle，然后看了看里面的题，感觉挺新颖的，就试了试。

题目如图：给了一个train.csv，现在预测test.csv里面的Passager是否幸存。train.csv里面包含的乘客信息有

PassagerId	乘客id
Survived	乘客是否幸存
Pclass	仓位
Name	乘客姓名
Sex	乘客性别
Age	乘客年龄
SibSp	船上是否有兄弟姐妹
Parch	穿上是否有父母子女
Ticket	船票信息
Fare	票价
Cabin	客舱
Embarked	上船地址

然后表里面的Sibsp，Parch，Name，PassagerId，Ticket，Cabin都是些数据无关的信息。

然后用到了随机森林算法。

#-*- coding:utf-8 -*-
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from subprocess import check_outputimport csv
import random as rnd
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.ensemble import RandomForestClassifier
from sklearn.cross_validation import cross_val_score
from sklearn.grid_search import GridSearchCV, RandomizedSearchCV
train_df = pd.read_csv('train.csv', header=0)
test_df = pd.read_csv('test.csv', header=0)
df = pd.concat([train_df, test_df])
df = df.reset_index()
df = df.drop('index',axis=1)
#移除index列
df = df.reindex_axis(train_df.columns,axis=1)
#填补合并之后的表中 属性是Age,Fare,Embarked为空的值
df['Age'][df['Age'].isnull()] = df['Age'].median()
df['Fare'][df['Fare'].isnull()] = df['Fare'].median()
df['Embarked'][df['Embarked'].isnull()] = df['Embarked'].mode().values
#将表中的Sex属性做映射
df['Sex'] = pd.factorize(df['Sex'])[0]
df['Embarked'] = pd.factorize(df['Embarked'])[0]
df['family_member'] = df['SibSp'] + df['Parch']
#移除表中的'Cabin','Ticke t','Name','SibSp','Parch','PassengerId'属性
d= df.drop(['Cabin','Ticke t','Name','SibSp','Parch','PassengerId'],axis=1)
survived_member = df[df['Survived'].notnull()].values
test_message = df[df['Survived'].isnull()].values
Y = survived_member[:, 0].astype(int)
#取servived属性不为空的属性的第一列
X = survived_member[:, 1:].astype(int)
#取servived属性不为空的出第一列之外的所有信息
result = RandomForestClassifier(n_estimators=1000, random_state=312, min_samples_leaf=3).fit(X, Y)
#随机森林算法
pre = result.predict(test_message[:, 1:]).astype(int)
Id = test_df['PassengerId']
result_csv = open('result1.csv',"w")
result_fd = csv.writer(result_csv)
result_fd.writerow(['PassengerId','Survived'])
result_fd.writerows(zip(Id,pre))
result_csv.close()