ML之DT：基于DT决策树算法(对比是否经特征筛选FS处理)对Titanic(泰坦尼克号)数据集进行二分类预测

输出结果

设计思路

核心代码

输出结果

初步处理后的 X_train： (984, 474)
(0, 0)   31.19418104265403
(0, 78)   1.0
(0, 82)   1.0
(0, 366)   1.0
(0, 391)   1.0
(0, 435)   1.0
(0, 437)   1.0
(0, 473)   1.0
(1, 0)   31.19418104265403
(1, 73)   1.0
(1, 79)   1.0
(1, 296)   1.0
(1, 389)   1.0
(1, 397)   1.0
(1, 436)   1.0
(1, 446)   1.0
(2, 0)   31.19418104265403
(2, 78)   1.0
(2, 82)   1.0
(2, 366)   1.0
(2, 391)   1.0
(2, 435)   1.0
(2, 437)   1.0
(2, 473)   1.0
(3, 0)   32.0
:   :
(980, 473)   1.0
(981, 0)   12.0
(981, 73)   1.0
(981, 81)   1.0
(981, 84)   1.0
(981, 390)   1.0
(981, 435)   1.0
(981, 436)   1.0
(981, 473)   1.0
(982, 0)   18.0
(982, 78)   1.0
(982, 81)   1.0
(982, 277)   1.0
(982, 390)   1.0
(982, 435)   1.0
(982, 437)   1.0
(982, 473)   1.0
(983, 0)   31.19418104265403
(983, 78)   1.0
(983, 82)   1.0
(983, 366)   1.0
(983, 391)   1.0
(983, 435)   1.0
(983, 436)   1.0
(983, 473)   1.0

经过FS处理后的 X_train_fs： (984, 94)
(0, 93)   1.0
(0, 85)   1.0
(0, 83)   1.0
(0, 76)   1.0
(0, 71)   1.0
(0, 27)   1.0
(0, 24)   1.0
(0, 0)   31.19418104265403
(1, 84)   1.0
(1, 74)   1.0
(1, 63)   1.0
(1, 25)   1.0
(1, 19)   1.0
(1, 0)   31.19418104265403
(2, 93)   1.0
(2, 85)   1.0
(2, 83)   1.0
(2, 76)   1.0
(2, 71)   1.0
(2, 27)   1.0
(2, 24)   1.0
(2, 0)   31.19418104265403
(3, 93)   1.0
(3, 85)   1.0
(3, 83)   1.0
:   :
(980, 24)   1.0
(980, 0)   31.19418104265403
(981, 93)   1.0
(981, 84)   1.0
(981, 83)   1.0
(981, 75)   1.0
(981, 28)   1.0
(981, 26)   1.0
(981, 19)   1.0
(981, 0)   12.0
(982, 93)   1.0
(982, 85)   1.0
(982, 83)   1.0
(982, 75)   1.0
(982, 26)   1.0
(982, 24)   1.0
(982, 0)   18.0
(983, 93)   1.0
(983, 84)   1.0
(983, 83)   1.0
(983, 76)   1.0
(983, 71)   1.0
(983, 27)   1.0
(983, 24)   1.0
(983, 0)   31.19418104265403

设计思路

核心代码

class SelectPercentile Found at: sklearn.feature_selection.univariate_selectionclass SelectPercentile(_BaseFilter):"""Select features according to a percentile of the highest scores.Read more in the :ref:`User Guide <univariate_feature_selection>`.Parameters----------score_func : callableFunction taking two arrays X and y, and returning a pair of arrays(scores, pvalues) or a single array with scores.Default is f_classif (see below "See also"). The default function onlyworks with classification tasks.percentile : int, optional, default=10Percent of features to keep.Attributes----------scores_ : array-like, shape=(n_features,)Scores of features.pvalues_ : array-like, shape=(n_features,)p-values of feature scores, None if `score_func` returned only scores.Notes-----Ties between features with equal scores will be broken in an unspecifiedway.See also--------f_classif: ANOVA F-value between label/feature for classification tasks.mutual_info_classif: Mutual information for a discrete target.chi2: Chi-squared stats of non-negative features for classification tasks.f_regression: F-value between label/feature for regression tasks.mutual_info_regression: Mutual information for a continuous target.SelectKBest: Select features based on the k highest scores.SelectFpr: Select features based on a false positive rate test.SelectFdr: Select features based on an estimated false discovery rate.SelectFwe: Select features based on family-wise error rate.GenericUnivariateSelect: Univariate feature selector with configurable mode."""def __init__(self, score_func=f_classif, percentile=10):super(SelectPercentile, self).__init__(score_func)self.percentile = percentiledef _check_params(self, X, y):if not 0 <= self.percentile <= 100:raise ValueError("percentile should be >=0, <=100; got %r" % self.percentile)def _get_support_mask(self):check_is_fitted(self, 'scores_')# Cater for NaNsif self.percentile == 100:return np.ones(len(self.scores_), dtype=np.bool)elif self.percentile == 0:return np.zeros(len(self.scores_), dtype=np.bool)scores = _clean_nans(self.scores_)treshold = stats.scoreatpercentile(scores, 100 - self.percentile)mask = scores > tresholdties = np.where(scores == treshold)[0]if len(ties):max_feats = int(len(scores) * self.percentile / 100)kept_ties = ties[:max_feats - mask.sum()]mask[kept_ties] = Truereturn mask