Day 081 — Titanicベースライン構築 — ロジスティック回帰でKaggle初提出

# 欠損補完
age_median = train['Age'].median()
df['Age'] = df['Age'].fillna(age_median)

# エンコード
df['Sex'] = df['Sex'].map({'male': 0, 'female': 1})

# Pipeline
pipeline = Pipeline([('scaler', StandardScaler()), ('lr', LogisticRegression())])

# CV
skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)
scores = cross_val_score(pipeline, X, y, cv=skf, scoring='accuracy')

import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import StratifiedKFold, cross_val_score
from sklearn.preprocessing import StandardScaler
from sklearn.pipeline import Pipeline

train = pd.read_csv('train.csv')
test  = pd.read_csv('test.csv')

def preprocess(df, age_med=None, embarked_mode=None):
    df = df.copy()
    if age_med is None: age_med = df['Age'].median()
    df['Age'] = df['Age'].fillna(age_med)
    if embarked_mode is None: embarked_mode = df['Embarked'].mode()[0]
    df['Embarked'] = df['Embarked'].fillna(embarked_mode)
    df['Sex']      = df['Sex'].map({'male': 0, 'female': 1})
    df['Embarked'] = df['Embarked'].map({'S': 0, 'C': 1, 'Q': 2})
    return df, age_med, embarked_mode

train, a, e  = preprocess(train)
test, _, _   = preprocess(test, a, e)

features = ['Pclass', 'Sex', 'Age', 'SibSp', 'Parch', 'Fare', 'Embarked']
X, y, X_test = train[features], train['Survived'], test[features]

pipeline = Pipeline([('sc', StandardScaler()), ('lr', LogisticRegression(max_iter=1000))])
skf      = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)
scores   = cross_val_score(pipeline, X, y, cv=skf, scoring='accuracy')
print(scores.mean())

pipeline.fit(X, y)
sub = pd.DataFrame({'PassengerId': test['PassengerId'], 'Survived': pipeline.predict(X_test)})
sub.to_csv('submission.csv', index=False)

import pandas as pd
import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import StratifiedKFold, cross_val_score
from sklearn.preprocessing import StandardScaler
from sklearn.pipeline import Pipeline

# データ読み込み（Kaggleカーネル内を想定）
train = pd.read_csv('/kaggle/input/titanic/train.csv')
test  = pd.read_csv('/kaggle/input/titanic/test.csv')

print(f"Train shape: {train.shape}")
print(f"Test shape:  {test.shape}")
print("\n欠損値確認:")
print(train.isnull().sum()[train.isnull().sum() > 0])

# ─── 前処理関数（data leakage 防止） ───
def preprocess(df, age_median=None, fare_median=None, embarked_mode=None):
    df = df.copy()
    if age_median is None:
        age_median = df['Age'].median()
    df['Age'] = df['Age'].fillna(age_median)
    if fare_median is None:
        fare_median = df['Fare'].median()
    df['Fare'] = df['Fare'].fillna(fare_median)
    if embarked_mode is None:
        embarked_mode = df['Embarked'].mode()[0]
    df['Embarked'] = df['Embarked'].fillna(embarked_mode)
    df['Sex']      = df['Sex'].map({'male': 0, 'female': 1})
    df['Embarked'] = df['Embarked'].map({'S': 0, 'C': 1, 'Q': 2})
    return df, age_median, fare_median, embarked_mode

train, age_med, fare_med, emb_mode = preprocess(train)
test,  _,       _,         _        = preprocess(test, age_med, fare_med, emb_mode)

# ─── 特徴量 / 目的変数 ───
features = ['Pclass', 'Sex', 'Age', 'SibSp', 'Parch', 'Fare', 'Embarked']
X      = train[features]
y      = train['Survived']
X_test = test[features]

# ─── Pipeline（標準化 + ロジスティック回帰） ───
pipeline = Pipeline([
    ('scaler', StandardScaler()),
    ('lr', LogisticRegression(max_iter=1000, random_state=42))
])

# ─── Stratified 5-Fold CV ───
skf    = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)
scores = cross_val_score(pipeline, X, y, cv=skf, scoring='accuracy')
print(f"\nCV Accuracy: {scores.mean():.4f} ± {scores.std():.4f}")
print(f"各Foldスコア: {np.round(scores, 4)}")

# ─── 最終学習 → 提出ファイル作成 ───
pipeline.fit(X, y)
preds = pipeline.predict(X_test)

submission = pd.DataFrame({
    'PassengerId': test['PassengerId'],
    'Survived':    preds
})
submission.to_csv('submission.csv', index=False)
print(f"\nsubmission.csv 作成完了: {len(submission)} 行")