python 利用库sklearn 中的 grid_search对svm 参数寻优（借鉴）

import time  
from sklearn import metrics  
import numpy as np  
import pickle 
from sklearn.tree import DecisionTreeClassifier
 
def svm_cross_validation(val_x, val_y):  
    from sklearn.grid_search import GridSearchCV  
    from sklearn.svm import SVC  
    model = SVC(kernel='rbf', probability=True)  
    param_grid = {'C': [ 1e-1, 1, 10], 'gamma': [0.001]}  
    grid_search = GridSearchCV(model, param_grid, n_jobs = 1, verbose=1)  
    grid_search.fit(train_x, train_y)  
    best_parameters = grid_search.best_estimator_.get_params()  
    for para, val in best_parameters.items():  
        print(para, val)  
    model = SVC(kernel='rbf', C=best_parameters['C'], gamma=best_parameters['gamma'], probability=True)  
    model.fit(train_x, train_y)