PCA & kNN(Titanic 데이터)

1. 데이터 준비

1) 이름 분리해서 title
2) title로 귀족과 평민 등급을 구별
3) gender, grade 컬럼 생성 → 머신러닝을 위해 sex, title 숫자로 표현
4) 결측치 제거

import pandas as pd

titanic = pd.read_excel(titanic_url)

# 이름 분리해서 title
import re
title = []

for idx, dataset in titanic.iterrows():
    title.append(re.search('\,\s\w+(\s\w+)?\.', dataset['name']).group()[2:-1])
titanic['title'] = title

# 귀족과 평민 등급을 구별
titanic['title'] = titanic['title'].replace('Mlle', 'Miss')
titanic['title'] = titanic['title'].replace('Ms', 'Miss')
titanic['title'] = titanic['title'].replace('Mme', 'Mrs')

Rare_f = ['Dona', 'Dr', 'Lady', 'the Countess']
Rare_m = ['Capt', 'Col', 'Don', 'Major', 'Rev', 'Sir', 'Jonkheer', 'Master']

for each in Rare_f:
    titanic['title'] = titanic['title'].replace(each, 'Rare_f')

for each in Rare_m:
    titanic['title'] = titanic['title'].replace(each, 'Rare_m')

# gender,grade 컬럼 -> 머신러닝을 위해 숫자로
from sklearn.preprocessing import LabelEncoder
le_gender = LabelEncoder()
le_gender.fit(titanic['sex'])
titanic['gender'] = le_gender.transform(titanic['sex'])

le_grade = LabelEncoder()
le_grade.fit(titanic['title'])
titanic['grade'] = le_grade.transform(titanic['title'])

# null이 아닌 데이터만
titanic = titanic[titanic['age'].notnull()]
titanic = titanic[titanic['fare'].notnull()]

2. 데이터 분리 & PCA 입력 출력 함수

import numpy as np
from sklearn.decomposition import PCA
from sklearn.model_selection import train_test_split

X = titanic[['pclass', 'age', 'sibsp', 'parch', 'fare', 'gender', 'grade']].astype(float)
y = titanic['survived']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=13)

def get_pca_data(ss_data, n_components=2):
    pca = PCA(n_components=n_components)
    pca.fit(ss_data)
    return pca.transform(ss_data), pca

def get_pd_from_pca(pca_data, col_num):
    cols = ['pca_' + str(n+1) for n in range(col_num)]
    return pd.DataFrame(pca_data, columns=cols)

def print_variance_ratio(pca, only_sum=False):
    if only_sum == False:
        print('variance_ration: ', pca.explained_variance_ratio_)
    print('sum of variance_ration: ', np.sum(pca.explained_variance_ratio_))

3. 함수 적용

• 2개 축으로 변환

pca_data, pca = get_pca_data(X_train, n_components=2) 
print_variance_ratio(pca)

# variance_ration:  [0.93577394 0.06326916]
# sum of variance_ration:  0.9990431009511275

4. 그래프

• 구분 안됨

import seaborn as sns

pca_pd = get_pd_from_pca(pca_data, 2)
pca_pd['survived'] = y_train

sns.pairplot(pca_pd, hue='survived', height=5, x_vars=['pca_1'], y_vars=['pca_2']) 

5. 함수 다시 적용

• 3개 축으로 변환

pca_data, pca = get_pca_data(X_train, n_components=3) 
print_variance_ratio(pca)

# variance_ration:  [9.35773938e-01 6.32691630e-02 4.00903990e-04]
# sum of variance_ration:  0.9994440049413547

6. pipline 구축

from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import accuracy_score
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler

estimators = [
    ('scaler', StandardScaler()),
    ('pca', PCA(n_components=3)),
    ('clf', KNeighborsClassifier(n_neighbors=20))
]

pipe = Pipeline(estimators)
pipe.fit(X_train, y_train)

pred = pipe.predict(X_test)
print(accuracy_score(y_test, pred)) # 0.7751196172248804

7. 디카프리오, 윈슬리 예측

# [['pclass', 'age', 'sibsp', 'parch', 'fare', 'gender', 'grade']]
dicaprio = np.array([[3, 18, 0, 0, 5, 1, 1]])
print('dicaprio: ', pipe.predict_proba(dicaprio)[0, 1]) # 0.1

winslet = np.array([[1, 16, 1, 1, 100, 0, 3]])
print('winslet: ', pipe.predict_proba(winslet)[0, 1]) # 0.85

Reference
1) 제로베이스 데이터스쿨 강의자료