gist for https://moritzkoerber.github.io/python/tutorial/2019/10/11/blogpost/

column-specific_preprocessing.py

import pandas as pd
from sklearn.compose import ColumnTransformer
from sklearn.impute import SimpleImputer
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import classification_report
from sklearn.model_selection import GridSearchCV, RepeatedStratifiedKFold
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import OneHotEncoder, StandardScaler

titanic = pd.read_csv('./titanic.csv')

categorical_features = ['pclass', 'sex', 'embarked']
categorical_transformer = Pipeline(
    [
        ('imputer_cat', SimpleImputer(strategy = 'constant', fill_value = 'missing')),
        ('onehot', OneHotEncoder(handle_unknown = 'ignore'))
    ]
)

numeric_features = ['age', 'sibsp', 'parch', 'fare']
numeric_transformer = Pipeline(
    [
        ('imputer_num', SimpleImputer(strategy = 'median')),
        ('scaler', StandardScaler())
    ]
)

preprocessor = ColumnTransformer(
    [
        ('categoricals', categorical_transformer, categorical_features),
        ('numericals', numeric_transformer, numeric_features)
    ],
    remainder = 'drop'
)

pipeline = Pipeline(
    [
        ('preprocessing', preprocessor),
        ('clf', LogisticRegression())
    ]
)

params = {
    'clf__solver': ['liblinear'],
    'clf__penalty': ['l1', 'l2'],
    'clf__C': [0.01, 0.1, 1, 10, 100],
    'clf__random_state': [42]
}

rskf = RepeatedStratifiedKFold(n_splits = 5, n_repeats = 2, random_state = 42)

cv = GridSearchCV(pipeline, params, cv = rskf, scoring = ['f1', 'accuracy'], refit = 'f1', n_jobs = -1)

X = titanic.drop('survived', axis = 1)
y = titanic.survived

cv.fit(X, y)

print(f'Best F1-score: {cv.best_score_:.3f}\n')
print(f'Best parameter set: {cv.best_params_}\n')
print(f'Scores: {classification_report(y, cv.predict(X))}')