python數(shù)據(jù)挖掘中比較實(shí)用的幾個(gè)特征選擇方法

import numpy as npfrom sklearn.tree import DecisionTreeClassifierfrom sklearn.externals.six import StringIOfrom sklearn import treeimport pydotplusclf = DecisionTreeClassifier()x = [[1,1,1,1,1,2,2,2,2,2,3,3,3,3,3],     [1,1,2,2,1,1,1,2,1,1,1,1,2,2,1],     [1,1,1,2,1,1,1,2,2,2,2,2,1,1,1],     [1,2,2,1,1,1,2,2,3,3,3,2,2,3,1]     ]y =  [1,1,2,2,1,1,1,2,2,2,2,2,2,2,1]x = np.array(x)x = np.transpose(x)clf.fit(x,y)print(clf.feature_importances_)feature_name = ['A1','A2','A3','A4']target_name = ['1','2']dot_data = StringIO()tree.export_graphviz(clf,out_file = dot_data,feature_names=feature_name,                     class_names=target_name,filled=True,rounded=True,                     special_characters=True)graph = pydotplus.graph_from_dot_data(dot_data.getvalue())graph.write_pdf("Tree.pdf")

from sklearn.linear_model import LinearRegressionimport numpy as npnp.random.seed(0)size = 5000#A dataset with 3 featuresX = np.random.normal(0, 1, (size, 3))#Y = X0 + 2*X1 + noiseY = X[:,0] + 2*X[:,1] + np.random.normal(0, 2, size)lr = LinearRegression()lr.fit(X, Y)#A helper method for pretty-printing linear modelsdef pretty_print_linear(coefs, names = None, sort = False):    if names == None:        names = ["X%s" % x for x in range(len(coefs))]    lst = zip(coefs, names)    if sort:        lst = sorted(lst,  key = lambda x:-np.abs(x[0]))    return " + ".join("%s * %s" % (round(coef, 3), name)                                   for coef, name in lst)
print "Linear model:", pretty_print_linear(lr.coef_)

from sklearn.cross_validation import ShuffleSplitfrom sklearn.metrics import r2_scorefrom collections import defaultdict
X = boston["data"]Y = boston["target"]
rf = RandomForestRegressor()scores = defaultdict(list)
#crossvalidate the scores on a number of different random splits of the datafor train_idx, test_idx in ShuffleSplit(len(X), 100, .3):    X_train, X_test = X[train_idx], X[test_idx]    Y_train, Y_test = Y[train_idx], Y[test_idx]    r = rf.fit(X_train, Y_train)    acc = r2_score(Y_test, rf.predict(X_test))    for i in range(X.shape[1]):        X_t = X_test.copy()        np.random.shuffle(X_t[:, i])        shuff_acc = r2_score(Y_test, rf.predict(X_t))        scores[names[i]].append((acc-shuff_acc)/acc)print "Features sorted by their score:"print sorted([(round(np.mean(score), 4), feat) for              feat, score in scores.items()], reverse=True)