classic.py

from operator import itemgetter

from sklearn.datasets import load_boston
from sklearn.ensemble import (
    BaggingRegressor,
    GradientBoostingRegressor,
    RandomForestRegressor,
)
from sklearn.metrics import make_scorer, mean_squared_error
from sklearn.model_selection import cross_validate
from sklearn.preprocessing import MinMaxScaler


def best_pick(report):
    idx = min(enumerate(report["test_score"]), key=itemgetter(1))[0]

    return dict(
        report,
        train_score_best=report["train_score"][idx],
        test_score_best=report["train_score"][idx],
    )


ds = load_boston()
points_x = MinMaxScaler().fit_transform(ds.data)

sizes = (1, 5, 10, 20, 50, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000)

print("BaggingRegressor")
print("================")
print()

print("forest_size,mse_training,mse_test")
for size in sizes:
    result = best_pick(
        cross_validate(
            BaggingRegressor(n_estimators=size, n_jobs=-1),
            points_x,
            ds.target,
            scoring=make_scorer(mean_squared_error),
            n_jobs=1,
            return_train_score=True,
            return_estimator=False,
        )
    )
    print(
        f"{size},{result['train_score_best']},{result['test_score_best']}", flush=True
    )

print()
print()
print("RandomForestRegressor")
print("=====================")
print()

print("forest_size,mse_training,mse_test")
for size in sizes:
    result = best_pick(
        cross_validate(
            RandomForestRegressor(n_estimators=size, bootstrap=False, n_jobs=-1),
            points_x,
            ds.target,
            scoring=make_scorer(mean_squared_error),
            n_jobs=1,
            return_train_score=True,
            return_estimator=False,
        )
    )
    print(
        f"{size},{result['train_score_best']},{result['test_score_best']}", flush=True
    )

print()
print()
print("GradientBoostingRegressor")
print("=========================")
print()

print("forest_size,mse_training,mse_test")
for size in sizes:
    result = best_pick(
        cross_validate(
            GradientBoostingRegressor(n_estimators=size, learning_rate=0.001),
            points_x,
            ds.target,
            scoring=make_scorer(mean_squared_error),
            n_jobs=-1,
            return_train_score=True,
            return_estimator=False,
        )
    )
    print(
        f"{size},{result['train_score_best']},{result['test_score_best']}", flush=True
    )