Tuning MLP by using Optuna.

mlp-optuna.py

import optuna
import sklearn
import sklearn.datasets
import sklearn.neural_network

def objective(trial):

    # ネットワーク構造の決定
    n_layers = trial.suggest_int('n_layers', 1, 4)
    layers = []
    for i in range(n_layers):
        layers.append(trial.suggest_int(f'n_units_{i}', 1, 100))

    # 学習・評価用データの取得
    mnist = sklearn.datasets.fetch_mldata('MNIST original')
    x_train, x_test, y_train, y_test = sklearn.model_selection.train_test_split(
        mnist.data, mnist.target)

    # モデルの学習
    clf = sklearn.neural_network.MLPClassifier(hidden_layer_sizes=tuple(layers))
    clf.fit(x_train, y_train)

    # 学習したモデルの評価
    return clf.score(x_test, y_test)

study = optuna.create_study(direction='maximize')
study.optimize(objective, n_trials=100)