scikit-learn 0.18に多層パーセプトロンが実装されてる、DeepLearningできそう

いつの間にか、scikit-learnが0.18にバージョンアップしていました。

http://scikit-learn.org/stable/whats_new.html

新しい機能の説明、Changelog > New features > Classifiers and Regressorsの中に、

The Gaussian Process module has been reimplemented and now offers classification and regression estimators through gaussian_process.GaussianProcessClassifier and gaussian_process.GaussianProcessRegressor. Among other things, the new implementation supports kernel engineering, gradient-based hyperparameter optimization or sampling of functions from GP prior and GP posterior. Extensive documentation and examples are provided. By Jan Hendrik Metzen.

Added new supervised learning algorithm: Multi-layer Perceptron (#3204) by Issam H. Laradji

Added linear_model.HuberRegressor, a linear model robust to outliers. (#5291) by Manoj Kumar.

Added the multioutput.MultiOutputRegressor meta-estimator. It converts single output regressors to multi-ouput regressors by fitting one regressor per output. By Tim Head.

とあります。Multi-layer Perceptron(多層パーセプトロン)で、活性化関数としてtanhやreluもあるので、いわゆるDeep Learningとして使えそうです。私はChainerで実行環境を構築してしまったので、なかなか使う機会がなさそうなのが残念です。もっと早く実装されていれば、こちらを使っていたかも。

1.17. Neural network models (supervised) — scikit-learn 0.18.1 documentation

MLPClassifierクラスで誤差逆伝搬で学習する多層パーセプトロンの実行例。

>>> from sklearn.neural_network import MLPClassifier
>>> X = [[0., 0.], [1., 1.]]
>>> y = [0, 1]
>>> clf = MLPClassifier(solver='lbfgs', alpha=1e-5,
...                     hidden_layer_sizes=(5, 2), random_state=1)
...
>>> clf.fit(X, y)                         
MLPClassifier(activation='relu', alpha=1e-05, batch_size='auto', beta_1=0.9, beta_2=0.999, early_stopping=False, epsilon=1e-08, hidden_layer_sizes=(5, 2), learning_rate='constant', learning_rate_init=0.001, max_iter=200, momentum=0.9,  nesterovs_momentum=True, power_t=0.5, random_state=1, shuffle=True,   solver='lbfgs', tol=0.0001, validation_fraction=0.1, verbose=False, warm_start=False)