[머신러닝] 캐글 타이타닉 분석코드 titanic.py

"""
Survived    Survival status    0 = No, 1 = Yes
PassengerId Problem delivered at test time
Pclass    Passenger Class  1 = 1st, 2 = 2nd, 3 = 3rd
Sex
Name
Age
SibSp    Number of Siblings/Spouses Aboard
Parch    Number of Parents/Children Aboard
Ticket    Ticket number
Fare    Passenger Fare
Cabin    Room number
boat - Lifeboat (if survived)
body - Body number (if did not survive and body was recovered)
Embarked    name of port on board C = Cherbourg, Q = Queenstown, S = Southampton
    Index(['PassengerId', 'Survived', 'Pclass', 'Name', 'Sex', 'Age', 'SibSp',
    'Parch', 'Ticket', 'Fare', 'Cabin', 'Embarked'],      dtype='object')
"""
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.model_selection import KFold
from sklearn.model_selection import cross_val_score
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
import matplotlib.pyplot as plt
import seaborn as sns
import warnings
warnings.simplefilter(action='ignore', category=FutureWarning)
 
 
class Titanic:
 
    context : str
    fname : str
    train : object
    test : object
    id : str
    model : object
    label : object
 
    @property
    def context(self) -> str: return self._context
    @context.setter
    def context(self, context): self._context = context
    @property
    def fname(self) -> str: return self._fname
    @fname.setter
    def fname(self, fname): self._fname = fname
    @property
    def train(self) -> str: return self._train
    @train.setter
    def train(self, train): self._train = train
    @property
    def test(self) -> str: return self._test
    @test.setter
    def test(self, test): self._test = test
    @property
    def id(self) -> str: return self._id
    @id.setter
    def id(self, id): self._id = id
    @property
    def label(self) -> str: return self._label
    @label.setter
    def label(self, label): self._label = label
    @property
    def model(self) -> str: return self._model
    @model.setter
    def model(self, model): self._model = model
 
    def new_file(self) -> str: return self._context + self._fname
 
    def new_dframe(self) -> object: return pd.read_csv(self.new_file())
 
    def modeling(self,this):
        print('1. Drop PassengerId, Cabin, Ticket')
        this = self.drop_feature(this, 'PassengerId')
        this = self.drop_feature(this, 'Cabin')
        this = self.drop_feature(this, 'Ticket')
        print('2. Embarked, Sex Nominal')
        this = self.embarked_nominal(this)
        this = self.sex_nominal(this)
        print('3. Fare Ordinal ')
        this = self.fare_ordinal(this)
        this = self.drop_feature(this, 'Fare')
        print('4. Title Nominal ')
        this = self.title_nominal(this)
        this = self.drop_feature(this, 'Name')
        print('5. Age Ordinal ')
        this = self.age_ordinal(this)
        print('6. Final Null Check')
        print('train null count \n {}'.format(this.train.isnull().sum()))
        print('test null count \n {}'.format(this.test.isnull().sum()))
        self.model = this.train.drop('Survived', axis=1)
        self.label = this.train['Survived']
        return this
 
    @staticmethod
    def age_ordinal(this)-> object:
        train = this.train
        test =  this.test
        train['Age'] = train['Age'].fillna(-0.5)
        test['Age'] = test['Age'].fillna(-0.5)
        bins = [-1, 0, 5, 12, 18, 24, 35, 60, np.inf]
        labels = ['Unknown', 'Baby', 'Child', 'Teenager','Student','Young Adult','Adult','Senior']
        train['AgeGroup'] = pd.cut(train['Age'], bins, labels=labels)
        test['AgeGroup'] = pd.cut(test['Age'], bins, labels=labels)
        age_title_mappeing = {
            0: 'Unknown', 1: 'Baby', 2: 'Child', 3: 'Teenager', 4: 'Student', 5: 'Young Adult', 6: 'Adult', 7: 'Senior'
        }
        for x in range(len(train['AgeGroup'])):
            if train['AgeGroup'][x] == 'Unknown':
                train['AgeGroup'][x] = age_title_mappeing[train['Title'][x]]
        for x in range(len(test['AgeGroup'])):
            if test['AgeGroup'][x] == 'Unknown':
                test['AgeGroup'][x] = age_title_mappeing[test['Title'][x]]
        age_mapping = {
            'Unknown': 0, 'Baby': 1, 'Child': 2, 'Teenager': 3, 'Student': 4,'Young Adult': 5, 'Adult': 6,'Senior':7
        }
        this.train['AgeGroup'] = train['AgeGroup'].map(age_mapping)
        this.test['AgeGroup'] = test['AgeGroup'].map(age_mapping)
        return this
 
    @staticmethod
    def title_nominal(this)-> object:
        combine = [this.train, this.test]
        for dataset in combine:
            dataset['Title'] = dataset.Name.str.extract('([A-Za-z])\.', expand=False)
        for dataset in combine:
            dataset['Title'] = dataset['Title'].replace(['Capt','Col','Don','Dr','Major','Rev','Jonkheer','Dona'], 'Rare')
            dataset['Title'] = dataset['Title'].replace(['Countess','Lady','Sir'], 'Royal')
            dataset['Title'] = dataset['Title'].replace(['Mile','Ms'], 'Miss')
        title_mapping = {"Mr": 1, "Miss": 2, "Mrs": 3, "Master": 4, "Royal": 5, "Rare": 6, "Mne": 7}
        for dataset in combine:
            dataset['Title'] = dataset['Title'].map(title_mapping)
            dataset['Title'] = dataset['Title'].fillna(0)
        return this
 
    @staticmethod
    def fare_ordinal(this) -> object:
        this.train['FareBand'] = pd.qcut(this.train['Fare'], 4, labels=[1,2,3,4])
        this.test['FareBand'] = pd.qcut(this.test['Fare'], 4, labels=[1,2,3,4])
        this.train = this.train.fillna({'FareBand': 1})
        this.test = this.test.fillna({'FareBand': 1})
        return this
 
    @staticmethod
    def drop_feature(this, feature) -> object:
        this.train = this.train.drop([feature], axis= 1)
        this.test = this.test.drop([feature], axis= 1)
        return this
 
    @staticmethod
    def embarked_nominal(this)-> object:
        this.train = this.train.fillna({"Embarked" : "S"})
        this.test = this.test.fillna({'Embarked': "S"})
        city_mapping = {"S": 1, "C": 2, "Q": 3}
        this.train['Embarked'] = this.train['Embarked'].map(city_mapping)
        this.test['Embarked'] = this.test['Embarked'].map(city_mapping)
        return this
 
    @staticmethod
    def sex_nominal(this) ->object:
        sex_mapping = {"male": 0, "female": 1}
        combine = [this.train, this.test]
        for dataset in combine:
            dataset['Sex'] = dataset['Sex'].map(sex_mapping)
        return this
 
    def learning(self, this):
        print('Decision Tree Accuracy {} % '.format(self.calculate_accuracy(this, DecisionTreeClassifier())))
        print('Random Forest Accuracy {} % '.format(self.calculate_accuracy(this, RandomForestClassifier())))
        print('KNN Accuracy {} % '.format(self.calculate_accuracy(this, KNeighborsClassifier())))
        print('Naive Bays Accuracy {} % '.format(self.calculate_accuracy(this, GaussianNB())))
        print('SVM Accuracy {} % '.format(self.calculate_accuracy(this, SVC())))
 
    @staticmethod
    def calculate_accuracy(this, classfier):
        score = cross_val_score(classfier,
                                this.model,
                                this.label,
                                cv= KFold(n_splits=10, shuffle= True, random_state=0),
                                n_jobs=1,
                                scoring='accuracy')
        return round(np.mean(score) * 100, 2)
 
class View:
    @staticmethod
    def plot_survived_dead(this):
        f, ax = plt.subplots(1, 2, figsize=(18, 8))
        this.train['Survived'].value_counts().plot.pie(explode=[0, 0.1],
                 autopct='%1.1f%%',
                 ax=ax[0],
                 shadow=True)
        ax[0].set_title('Survived')
        ax[1].set_ylabel('')
        sns.countplot('Survived', data=this.train, ax=ax[1])
        ax[1].set_title('Survived')
        plt.show()
 
    @staticmethod
    def plot_sex(this):
        f, ax = plt.subplots(1, 2, figsize=(18, 8))
        this.train['Survived'][this.train['Sex'] == 'male'].value_counts()\
            .plot.pie(explode=[0, 0.1],autopct='%1.1f%%',ax=ax[0],shadow=True)
        this.train['Survived'][this.train['Sex'] == 'female'].value_counts()\
            .plot.pie(explode=[0, 0.1],autopct='%1.1f%%',ax=ax[1],shadow=True)
        ax[0].set_title('Male')
        ax[1].set_title('Femail')
        plt.show()
 
    @staticmethod
    def bar_chart(this, feature):
        survived = this.train[this.train['Survived']==1][feature].value_counts()
        dead = this.train[this.train['Survived']==0][feature].value_counts()
        df = pd.DataFrame([survived, dead])
        df.index = ['survived', 'dead']
        df.plot(kind='bar', stacked=True, figsize=(110,1))
        plt.show()
 
if __name__ == '__main__':
    def print_menu():
        print('0. Exit')
        print('1. Create Model')
        print('2. Data Visualize')
        print('3. Modeling')
        print('4. Learning')
        print('5. Submit')
        return input('Choose One\n')
 
    this = Titanic()
    view = View()
 
    while 1:
        menu = print_menu()
        print(f'Menu : {menu} ')
        if menu == '0':
            print('Stop')
        if menu == '1':
            this.context = './data/'
            this.fname = 'train.csv'
            this.new_file()
            this.train = this.new_dframe()
            this.fname = 'test.csv'
            this.new_file()
            this.test = this.new_dframe()
            this.id = this.test['PassengerId']
        if menu == '2':
            view.plot_survived_dead(this)
            # view.plot_sex(this)
            # view.bar_chart(this, 'Pclass')
        if menu == '3':
            this = this.modeling(this)
 
        if menu == '4':
            this.learning(this)
        if menu == '5':
            clf = SVC()
            clf.fit(this.model, this.label)
            prediction = clf.predict(this.test)
            submission = pd.DataFrame(
                {'PassengerId': this.id, 'Survived': prediction}
            )
            submission.to_csv('./data/submission.csv', index=False)