Main Content

本页翻译不是最新的。点击此处可查看最新英文版本。

广义加性模型

用于回归的由一元和二元形状函数组成的可解释模型

使用 fitrgam 拟合用于回归的广义加性模型。

广义加性模型 (GAM) 是一种可解释的模型，它使用预测变量的一元和二元形状函数之和来解释响应变量。fitrgam 使用提升树作为每个预测变量以及可选的每对预测变量的形状函数；因此，该函数可以捕获预测变量和响应变量之间的非线性关系。由于单个形状函数对预测（响应值）的贡献相互独立，因此该模型易于解释。

对象

RegressionGAMGeneralized additive model (GAM) for regression (自 R2021a 起)
CompactRegressionGAMCompact generalized additive model (GAM) for regression (自 R2021a 起)
RegressionPartitionedGAMCross-validated generalized additive model (GAM) for regression (自 R2021a 起)

函数

全部展开

fitrgamFit generalized additive model (GAM) for regression (自 R2021a 起)
compactReduce size of machine learning model
crossvalCross-validate machine learning model
templateGAMGeneralized additive model (GAM) learner template (自 R2023b 起)
addInteractionsAdd interaction terms to univariate generalized additive model (GAM) (自 R2021a 起)
resumeResume training of generalized additive model (GAM) (自 R2021a 起)
limeLocal interpretable model-agnostic explanations (LIME) (自 R2020b 起)
partialDependenceCompute partial dependence (自 R2020b 起)
permutationImportancePredictor importance by permutation (自 R2024a 起)
plotLocalEffectsPlot local effects of terms in generalized additive model (GAM) (自 R2021a 起)
plotPartialDependenceCreate partial dependence plot (PDP) and individual conditional expectation (ICE) plots
shapleyShapley values (自 R2021a 起)
predictPredict responses using generalized additive model (GAM) (自 R2021a 起)
lossRegression loss for generalized additive model (GAM) (自 R2021a 起)
resubPredictPredict responses for training data using trained regression model
resubLossResubstitution regression loss
kfoldPredictPredict responses for observations in cross-validated regression model
kfoldLossLoss for cross-validated partitioned regression model
kfoldfunCross-validate function for regression

主题