DataAnalysis Kaggle Titanic with Reference

'''
패키지 라이브러리 (pandas, numpy) import
'''
import pandas as pd
import numpy as np

train_original = pd.read_csv('input/train.csv')
test_original = pd.read_csv('input/test.csv')

train = train_original.copy()
test = test_original.copy()

# 1단계 데이터 살펴보기
print("SHAPE: ", train.shape)
print(train.isnull().sum())
train.head(3)

train.info()

print("SHAPE: ", test.shape)
print(test.isnull().sum())
test.head(3)

test.info()

# VISUALIZATION 라이브러리리 호출
import matplotlib.pyplot as plt
%matplotlib inline
import seaborn as sns
sns.set() #Setting Seaborn default for plots

def bar_chart(df, feature):
    survived = df.loc[df['Survived']==1,feature].value_counts()
    dead = df.loc[df['Survived']==0, feature].value_counts()
    mat = pd.DataFrame([survived, dead])
    mat.index = ['Survived','Dead']
    mat.plot(kind='bar', stacked = True, figsize=(10,5))

bar_chart(train, 'Sex') #남자가 많이 죽고, 여자가 많이 살았네 (strongly)
bar_chart(train, 'Pclass') # 3등급이 많이 죽긴 하였네 
bar_chart(train, 'SibSp') # 가족이 없는 경우네는 많이 죽긴 하였네 (weakly)
bar_chart(train, 'Parch') 
bar_chart(train, 'Embarked') 

# 1등급/2등급/3등급 승객의 탑승지 bar_chart
pclass1 = train.loc[train['Pclass']==1, 'Embarked'].value_counts()
pclass2 = train.loc[train['Pclass']==2, 'Embarked'].value_counts()
pclass3 = train.loc[train['Pclass']==3, 'Embarked'].value_counts()
sset = pd.DataFrame([pclass1, pclass2, pclass3])
sset.index = ['Pclass1','Pclass2', 'Pclass3']
sset.plot(kind='bar', stacked = True, figsize=(10,5))

# 이름에서 성별 정보등을 Title이라는 키로 추출하기
train_test_data = [train, test]

for dataset in train_test_data:
    dataset['Title']=dataset['Name'].str.extract('([A-Za-z]+)\.', expand=False)

train['Title'].value_counts()

train.isnull().sum()

title_mapping = {"Mr": 0, "Miss": 1, "Mrs": 2, "Master":3, "Dr": 3, "Rev": 3, "Col": 3 , "Major":3, "Mlle":3, "Countess":3, "Ms": 3, "Lady": 3 , "Jonkheer": 3, "Don": 3, "Dona": 3 , "Mme":3, "Capt": 3, "Sir":3}

for dataset in train_test_data:
    dataset['Title']=dataset['Title'].map(title_mapping)

bar_chart(train, 'Title')
train['Title'].value_counts() #보면 Mr(0)는 확실히 많이 죽었음
train.drop('Name', axis=1, inplace=True)
test.drop('Name', axis=1, inplace=True)

sex_mapping = {'male': 0, "female": 1}

for dataset in train_test_data:
    dataset['Sex'] = dataset['Sex'].map(sex_mapping)

# 결측데이터 채우기 (물론 버려도 됨)
# 여기서는 Mr / Ms / Mrs 그룹들의 평균 나이를 결측데이터로 넣어줄것임
train['Age'].fillna(train.groupby('Title')['Age'].transform('median'), inplace=True)
test['Age'].fillna(test.groupby('Title')['Age'].transform('median'), inplace=True)

train.isnull().sum()

def facet_plot(df,feature, range_opt=None):
    facet = sns.FacetGrid(df, hue='Survived', aspect=4)
    facet.map(sns.kdeplot, feature, shade = True)

    if not range_opt:
        facet.set(xlim=(0, train[feature].max()))
    else:
        facet.set(xlim=range_opt)
    facet.add_legend()

    plt.show()

facet_plot(train,'Age')
facet_plot(train,'Age', [0,20])
facet_plot(train,'Age', [20,35]) # 20~35세는 많이 죽은 것을 볼 수 있네
facet_plot(train,'Age', [35,80])

for dataset in train_test_data:
    dataset.loc[ dataset['Age'] <= 16 , 'Age'] =0
    dataset.loc[(dataset['Age'] > 16) & (dataset['Age'] <=26), 'Age'] = 1
    dataset.loc[(dataset['Age'] > 26) & (dataset['Age'] <=36), 'Age'] = 2
    dataset.loc[(dataset['Age'] > 36) & (dataset['Age'] <=62), 'Age'] = 3
    dataset.loc[ dataset['Age'] > 62  , 'Age'] = 4

# Embarked 결측데이터 채우기
# 보통 S에서 탔더라고,S로 채우자

embarked_mapping={"S": 0, "C": 1, "Q": 2}

for dataset in train_test_data:
    dataset['Embarked']=dataset['Embarked'].fillna('S')
    dataset['Embarked'] = dataset['Embarked'].map(embarked_mapping)

facet_plot(train, 'Fare')
facet_plot(train, 'Fare', [0,30]) #이 구간은 많이 죽음
facet_plot(train, 'Fare', [30,150]) # 이 이후부터는 많이 살았음

# binning
for dataset in train_test_data:
    dataset.loc[dataset['Fare'] <= 17, 'Fare'] =0 
    dataset.loc[(dataset['Fare'] > 17) & (dataset['Fare'] <=30), 'Fare'] =1 
    dataset.loc[(dataset['Fare'] > 30) & (dataset['Fare'] <=100), 'Fare'] =2
    dataset.loc[dataset['Fare'] > 100, 'Fare'] =3  
    

train['Fare'].value_counts()

facet_plot(train,'Fare')

train['Cabin'].value_counts()

cabin_mapiing={"A": 0, "B": 0.4 , "C": 0.8, "D": 1.2, "E": 1.6, "F": 2, "G": 2.4, "T": 2.8}
for dataset in train_test_data:
    dataset['Cabin'] =dataset['Cabin'].str[:1]
    dataset['Cabin'] = dataset['Cabin'].map(cabin_mapiing)

train['Cabin'].fillna( train.groupby('Pclass')['Cabin'].transform('median'), inplace=True)
test['Cabin'].fillna( test.groupby('Pclass')['Cabin'].transform('median'), inplace=True)

test['Cabin'].value_counts()

test.isnull().sum()

train['Fare'].fillna(train.groupby('Pclass')['Fare'].transform('median'), inplace=True)
test['Fare'].fillna(train.groupby('Pclass')['Fare'].transform('median'), inplace=True)

train['FamilySize'] = train['SibSp'] + train['Parch']+1
test['FamilySize'] = test['SibSp'] + test['Parch'] +1

facet_plot(train, "FamilySize")

family_mapping = {1: 0, 2: 0.4, 3: 0.8, 4: 1.2, 5:1.6, 6:2, 7:2.4, 8:2.8, 9:3.2, 10:3.6, 11:4}

for dataset in train_test_data:
    dataset['FamilySize'] = dataset['FamilySize'].map(family_mapping)

features_drop = ['Ticket', 'SibSp', 'Parch', 'PassengerId']
train = train.drop(features_drop, axis= 1)
test = test.drop(features_drop, axis=1)

train_data = train.drop('Survived', axis =1)
target = train['Survived']

train_data.shape, target.shape, test.shape

train_data.head(3)

from sklearn.neighbors import KNeighborsClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn.svm import SVC

from sklearn.model_selection import KFold
from sklearn.model_selection import cross_val_score
k_fold = KFold(n_splits = 10, shuffle = True, random_state = 0)

clf = KNeighborsClassifier(n_neighbors=13)
scoring = 'accuracy'
score = cross_val_score(clf, train_data, target, cv= k_fold, n_jobs =1 , scoring=scoring)
print(np.mean(score))

clf = RandomForestClassifier(n_estimators=13)
score = cross_val_score(clf, train_data, target, cv= k_fold, n_jobs =1 , scoring=scoring)
print(np.mean(score))

clf = GaussianNB()
score = cross_val_score(clf, train_data, target, cv= k_fold, n_jobs =1 , scoring=scoring)
print(np.mean(score))

clf = SVC()
score = cross_val_score(clf, train_data, target, cv= k_fold, n_jobs =1 , scoring=scoring)
print(np.mean(score))

clf = SVC() #SVM 모델이 정확도가 제일 좋으니까 83%로,, 이걸로 예측을 하자!
clf.fit(train_data, target)

prediction = clf.predict(test)

# 출력하기
submission = pd.DataFrame\
({"PassengerId": test_original['PassengerId'],
  "Survived": prediction })

submission.to_csv('submission_wook.csv',index=False)

submission = pd.read_csv('submission_wook.csv')
submission.head()