scikit-learn の使い方 (その3)

こちらで行ったことを Kaggle で行いました。スコアは、0.77511 でした。
scikit-learn の使い方 (その2)

[1]

import numpy as np
import pandas as pd

from sklearn.preprocessing import LabelEncoder
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC

[2]

train_df = pd.read_csv("../input/titanic/train.csv", header=0)
test_df = pd.read_csv("../input/titanic/test.csv", header=0)
ids = test_df["PassengerId"].values


train_df = train_df[['Survived', 'Pclass', 'Sex', 'Fare']]
train_df['Fare'] = train_df['Fare'].fillna(train_df['Fare'].median())




encoder_sex = LabelEncoder()
train_df['Sex'] = encoder_sex.fit_transform(train_df['Sex'].values)
standard = StandardScaler()
train_df_std = pd.DataFrame(standard.fit_transform(train_df[['Pclass', 'Fare']]), columns=['Pclass', 'Fare'])

train_df['Pclass'] = train_df_std['Pclass']
train_df['Fare'] = train_df_std['Fare']

train_x = train_df.drop('Survived',axis = 1)
train_y = train_df.Survived

[3]

test_df = test_df[['Pclass', 'Sex', 'Fare']]
test_df['Fare'] = test_df['Fare'].fillna(test_df['Fare'].median())
encoder_sex = LabelEncoder()
test_df['Sex'] = encoder_sex.fit_transform(test_df['Sex'].values)
standard = StandardScaler()
test_df_std = pd.DataFrame(standard.fit_transform(test_df[['Pclass', 'Fare']]), columns=['Pclass', 'Fare'])

test_df['Pclass'] = test_df_std['Pclass']
test_df['Fare'] = test_df_std['Fare']
test_x = test_df

[4]

model = SVC(random_state=1,max_iter=5000)
model.fit(train_x, train_y)

pred = model.predict(test_x)

[5]

file_submit = "titanic_logistic_regression.csv"
#
dft = pd.DataFrame({'PassengerId': ids, 'Survived': pred})
dft.to_csv(file_submit,index=False)
dft.head()