python中Scikit-learn库的高级特性和实践分享

from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC
pipe = Pipeline([
    ('scaler', StandardScaler()),
    ('svm', SVC())
])
pipe.fit(X_train, y_train)
predictions = pipe.predict(X_test)

from sklearn.metrics import make_scorer
def custom_loss_func(y_true, y_pred):
    # Insert your own calculation here
    return loss_value
my_scorer = make_scorer(custom_loss_func, greater_is_better=False)
grid_search = GridSearchCV(estimator, param_grid, scoring=my_scorer)

from sklearn.externals import joblib
# Save the model
joblib.dump(clf, 'model.pkl')
# Load the model
clf = joblib.load('model.pkl')

from sklearn.pipeline import FeatureUnion
from sklearn.decomposition import PCA
from sklearn.decomposition import KernelPCA
combined_features = FeatureUnion([
    ("pca", PCA(n_components=3)),
    ("kernel_pca", KernelPCA(n_components=3)),
])
X_features = combined_features.fit(X, y).transform(X)