머신러닝 파이프라인

엔드 투 엔드 머신러닝 워크플로를 파이프라인으로 실행 및 배포

파이프라인은 데이터 처리 워크플로의 여러 단계를 결합하고 구성합니다. 머신러닝에서는 데이터 준비, 특징 엔지니어링, 특징 선택, 모델링 및 후처리 등이 이러한 단계에 속합니다. 파이프라인을 실행하면 데이터가 파이프라인을 통과할 때 각 단계가 데이터에 적용됩니다. 데이터 전처리 전용 또는 특징 엔지니어링 전용 파이프라인을 생성할 수 있습니다. 또는 데이터 전처리, 특징 엔지니어링, 분류 또는 회귀, 추론을 위한 여러 단계가 함께 포함된 하나의 머신러닝 파이프라인을 생성할 수 있습니다.

객체

모두 확장

파이프라인 객체

LearningPipeline Machine learning pipeline (R2026a 이후)

데이터 전처리 컴포넌트

`equalWidthBinnerComponent`	Pipeline component for grouping data into equal-width bins (R2026a 이후)
`frequencyEncoderComponent`	Pipeline component for frequency encoding categorical variables (R2026a 이후)
`kmeansEncoderComponent`	Pipeline component for feature extraction using k-means clustering (R2026a 이후)
`normalizerComponent`	Pipeline component for normalizing data (R2026a 이후)
`observationImputerComponent`	Pipeline component for imputing missing values (R2026a 이후)
`observationRemoverComponent`	Pipeline component for removing observations (R2026a 이후)
`oneHotEncoderComponent`	Pipeline component for encoding categorical data into one-hot vectors (R2026a 이후)
`outlierImputerComponent`	Pipeline component for imputing outlier values (R2026a 이후)
`outlierRemoverComponent`	Pipeline component for removing outlier values (R2026a 이후)
`pcaComponent`	Pipeline component for principal component analysis (PCA) (R2026a 이후)
`quantileBinnerComponent`	Pipeline component for binning data based on quantiles (R2026a 이후)
`ricaComponent`	Pipeline component for feature extraction using reconstruction independent component analysis (RICA) (R2026a 이후)
`sparseFilterComponent`	Pipeline component for feature extraction using sparse filtering (R2026a 이후)

특징 선택 및 엔지니어링 컴포넌트

`featureSelectionClassificationANOVAComponent`	Pipeline component for performing feature selection using ANOVA algorithm (R2026a 이후)
`featureSelectionClassificationChi2Component`	Pipeline component for performing feature selection using chi-square tests (R2026a 이후)
`featureSelectionClassificationKruskalWallisComponent`	Pipeline component for performing feature selection using Kruskal-Wallis test (R2026a 이후)
`featureSelectionClassificationMRMRComponent`	Pipeline component for performing MRMR feature selection in classification workflow (R2026a 이후)
`featureSelectionClassificationNCAComponent`	Pipeline component for performing feature selection using neighborhood component analysis (NCA) for classification (R2026a 이후)
`featureSelectionClassificationReliefFComponent`	Pipeline component for performing feature selection using ReliefF algorithm (R2026a 이후)
`featureSelectionRegressionFTestComponent`	Pipeline component for performing feature selection using F-tests (R2026a 이후)
`featureSelectionRegressionMRMRComponent`	Pipeline component for performing MRMR feature selection in regression workflow (R2026a 이후)
`featureSelectionRegressionNCAComponent`	Pipeline component for performing feature selection using neighborhood component analysis (NCA) for regression (R2026a 이후)
`featureSelectionRegressionReliefFComponent`	Pipeline component for performing feature selection using RReliefF algorithm (R2026a 이후)
`variableSelectorComponent`	Pipeline component for manual variable selection (R2026a 이후)

지도 학습 컴포넌트

분류 컴포넌트

`classificationDiscriminantComponent`	Pipeline component for discriminant analysis classification (R2026a 이후)
`classificationECOCComponent`	Pipeline component for multiclass classification using error-correcting output codes (ECOC) model (R2026a 이후)
`classificationEnsembleComponent`	Pipeline component for ensemble classification (R2026a 이후)
`classificationGAMComponent`	Pipeline component for binary classification using generalized additive model (GAM) (R2026a 이후)
`classificationKernelComponent`	Pipeline component for classification using Gaussian kernel with random feature expansion (R2026a 이후)
`classificationKNNComponent`	Pipeline component for classification using k-nearest neighbor model (R2026a 이후)
`classificationLinearComponent`	Pipeline component for binary classification of high-dimensional data using linear model (R2026a 이후)
`classificationNaiveBayesComponent`	Pipeline component for multiclass classification using naive Bayes model (R2026a 이후)
`classificationNeuralNetworkComponent`	Pipeline component for classification using neural network model (R2026a 이후)
`classificationSVMComponent`	Pipeline component for one-class and binary classification using SVM classifier (R2026a 이후)
`classificationTreeComponent`	Pipeline component for multiclass classification using binary decision trees (R2026a 이후)

회귀 컴포넌트

`regressionEnsembleComponent`	Pipeline component for regression using ensemble of learners (R2026a 이후)
`regressionGAMComponent`	Pipeline component for generalized additive model (GAM) for regression (R2026a 이후)
`regressionGPComponent`	Pipeline component for Gaussian process regression (GPR) (R2026a 이후)
`regressionLinearComponent`	Pipeline component for regression of high-dimensional data using a linear model (R2026a 이후)
`regressionKernelComponent`	Pipeline component for regression using Gaussian kernel model (R2026a 이후)
`regressionNeuralNetworkComponent`	Pipeline component for regression using neural network model (R2026a 이후)
`regressionSVMComponent`	Pipeline component for regression using a support vector machine (SVM) model (R2026a 이후)
`regressionTreeComponent`	Pipeline component for regression using binary decision trees (R2026a 이후)

사용자 지정 컴포넌트

functionComponent Pipeline component for custom function (R2026a 이후)

함수

모두 확장

생성 및 수정

자동 연결

`series`	Connect components in series to create pipeline (R2026a 이후)
`parallel`	Connect components or pipelines in parallel to create pipeline (R2026a 이후)
`insert`	Insert component or pipeline into existing pipeline (R2026a 이후)
`replace`	Replace existing pipeline component with new component (R2026a 이후)

수동 연결

`add`	Add new component or pipeline to existing pipeline (R2026a 이후)
`remove`	Remove existing components or pipelines from pipeline (R2026a 이후)
`connect`	Create connections between pipeline components (R2026a 이후)
`disconnect`	Remove connections between ports in pipeline (R2026a 이후)

계층 구조

expand Expand subpipelines in pipeline (R2026a 이후)

실행

`learn`	Initialize and evaluate pipeline or component (R2026a 이후)
`run`	Execute pipeline or component for inference after learning (R2026a 이후)
`prune`	Remove unnecessary components and dependencies from pipeline (R2026a 이후)
`reset`	Reset pipeline or component (R2026a 이후)
`crossvalidate`	Cross-validate pipeline (R2026a 이후)

시각화와 설명

`view`	View diagram of pipeline inputs, outputs, components, and connections (R2026a 이후)
`describe`	Display summary of pipeline components (R2026a 이후)

배포

package Create deployable archive or standalone application from pipeline (R2026a 이후)

도움말 항목

Machine Learning Pipeline Phases
Understand the learn and run pipeline phases for local and deployed execution.