scikit-learn Cookbook , Second Edition - Second Edition

How to do it...

1. Create a OneVsRestClassifier within a pipeline and support vector regression (SVR) from sklearn.svm:

from sklearn.svm import SVRfrom sklearn.pipeline import Pipelinefrom sklearn.preprocessing import StandardScalerfrom sklearn.multiclass import OneVsRestClassifier svm_est = Pipeline([('scaler',StandardScaler()),('svc',OneVsRestClassifier(SVR()))])

2. Create a parameter grid:

Cs = [0.001, 0.01, 0.1, 1]gammas = [0.001, 0.01, 0.1] param_grid = dict(svc__estimator__gamma=gammas, svc__estimator__C=Cs)

3. Perform a randomized search of the best hyperparameters, C and gamma:

from sklearn.model_selection import RandomizedSearchCVfrom sklearn.model_selection import StratifiedShuffleSplit rand_grid = RandomizedSearchCV(svm_est, param_distributions=param_grid, ...