Pipeline, ColumnTransformer, Pandas

snippet.py

# See https://machinelearningmastery.com/columntransformer-for-numerical-and-categorical-data/
from sklearn.linear_model import LinearRegression
from sklearn.compose import ColumnTransformer
from sklearn.impute import SimpleImputer
from sklearn.pipeline import Pipeline
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder, StandardScaler
from sklearn import set_config
import seaborn as sns
set_config(display='diagram')   

df = sns.load_dataset('penguins')
df.dropna(subset = ["body_mass_g"], inplace=True)
X = df.drop("body_mass_g", axis=1)
y = df["body_mass_g"]

numerical_ix = X.select_dtypes(include=['int64', 'float64']).columns
categorical_ix = X.select_dtypes(include=['object', 'bool']).columns

numeric_transformer = Pipeline(steps=[
    ('impute', SimpleImputer(strategy='median')),
    ('scaler', StandardScaler())
])

categorical_transformer = Pipeline(steps=[
    ('impute', SimpleImputer(strategy='constant', fill_value="unknown")),
    ('onehot2', OneHotEncoder(handle_unknown='ignore'))
])

preprocessor = ColumnTransformer(transformers=[
    ('num', numeric_transformer, numerical_ix),
    ('cat', categorical_transformer, categorical_ix),
])

# 全体のパイプラインの作成
pipe = Pipeline([
    ("preprocessor", preprocessor),  
    ("model", LinearRegression())
])

# show
pipe

# Train/Test
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=123)
pipe.fit(X_train, y_train)
print(pipe.score(X_test, y_test))
print('Intercept: \n', pipe["model"].intercept_)
print('Coefficients: \n', pipe["model"].coef_)