预测象征
估计classificati预测的重要性on ensemble of decision trees
句法
一世mp = predictorImportance(ens)
[[一世mp,ma] = predictorImportance(ens)
Description
computes estimates of predictor importance for一世mp
=预测象征((ens
)ens
by summing these estimates over all weak learners in the ensemble.一世mp
has one element for each input predictor in the data used to train this ensemble. A high value indicates that this predictor is important forens
。
[[
返回a一世mp
,,,,ma
] = predictorImportance(ens
)p
-by-p
矩阵具有相关性的预测度量p
predictors, when the learners inens
contain surrogate splits. See更多关于。
我nput Arguments
|
一个classification ensemble of decision trees, created by |
输出参数
|
一个row vector with the same number of elements as the number of predictors (columns) in |
|
一个 |
eXamples
更多关于
一个lgorithms
元素ma(i,j)
是在预测器上对替代分裂平均的关联的预测度量j
for which predictor一世
是最佳的拆分预测变量。通过求和预测器上的最佳分割的相关性预测度量的正值来计算该平均值一世
并在预测器上替代拆分j
and dividing by the total number of optimal splits on predictor一世
,包括拆分预测指标之间关联的预测度量一世
andj
负面。