Simulink에서의 머신러닝
R2026bSimulink를 사용한 머신러닝 워크플로 확장
Statistics and Machine Learning Toolbox™에 포함되어 있는 Statistics and Machine Learning 블록 라이브러리의 블록을 사용하여 Simulink® 모델에서 머신러닝 기능을 구현합니다. 이 툴박스는 다음과 같은 워크플로를 수행하는 블록을 제공합니다.
분류 예측 블록 또는 회귀 예측 블록을 사용하여, 훈련된 분류 모델 객체 또는 회귀 모델 객체를 Simulink에 가져옵니다.
분류 학습기 앱 또는 회귀 학습기 앱에서 머신러닝 모델을 훈련시키고 해당 모델을 Simulink로 내보냅니다.
Simulink에서 점진적 학습 블록을 사용하여 머신러닝 모델에서 실시간으로 드리프트를 지속적으로 업데이트하고 모니터링합니다.
데이터에서 쿼리 점에 대한 최근접이웃을 찾고 KNN Search 블록을 사용하여 Simulink에서 군집 분석을 수행합니다.
Python 연동실행 블록을 사용하여, 훈련된 Python® 머신러닝 모델을 Simulink에서 연동실행합니다.
블록
 | ClassificationECOC Predict | Classify observations using error-correcting output codes (ECOC) classification model (R2023a 이후) |
 | ClassificationEnsemble Predict | Classify observations using ensemble of decision trees |
 | ClassificationKNN Predict | Classify observations using nearest neighbor classification model (R2022b 이후) |
 | ClassificationLinear Predict | Classify observations using linear classification model (R2023a 이후) |
 | ClassificationNaiveBayes Predict | Classify observations using naive Bayes model (R2023b 이후) |
 | ClassificationNeuralNetwork Predict | Classify observations using neural network classification model |
 | ClassificationSVM Predict | Classify observations using support vector machine (SVM) classifier for one-class and binary classification |
 | ClassificationTree Predict | Classify observations using decision tree classifier |
 | ClassificationDiscriminant Predict | Classify observations using discriminant analysis model (R2024a 이후) |
 | ClassificationKernel Predict | Classify observations using Gaussian kernel classifier for binary classification (R2024b 이후) |
 | ClassificationXGBoost Predict | Classify observations using XGBoost classification model (R2026b 이후) |
 | RegressionEnsemble Predict | Predict responses using ensemble of decision trees for regression |
 | RegressionGP Predict | Predict responses using Gaussian process (GP) regression model (R2022a 이후) |
 | RegressionLinear Predict | 선형 회귀 모델을 사용하여 응답 예측 (R2023a 이후) |
 | RegressionNeuralNetwork Predict | 신경망 회귀 모델을 사용하여 응답 변수 예측 |
 | RegressionSVM Predict | 서포트 벡터 머신(SVM) 회귀 모델을 사용하여 응답 변수 예측 |
 | RegressionTree Predict | 회귀 트리 모델을 사용하여 응답 변수 예측 |
 | RegressionKernel Predict | Predict responses using Gaussian kernel regression model (R2024b 이후) |
 | RegressionXGBoost Predict | Predict responses using XGBoost regression model (R2026b 이후) |
 | IncrementalClassificationLinear Predict | Classify observations using incremental linear classification model (R2023b 이후) |
 | IncrementalClassificationLinear Fit | Fit incremental linear binary classification model (R2023b 이후) |
 | IncrementalRegressionLinear Predict | Predict responses using incremental linear regression model (R2023b 이후) |
 | IncrementalRegressionLinear Fit | Fit incremental linear regression model (R2023b 이후) |
 | IncrementalClassificationECOC Fit | Fit incremental ECOC classification model (R2024a 이후) |
 | IncrementalClassificationECOC Predict | Classify observations using incremental ECOC classification model (R2024a 이후) |
 | IncrementalClassificationKernel Fit | Fit incremental kernel classification model (R2024b 이후) |
 | IncrementalClassificationKernel Predict | Classify observations using incremental kernel classification model (R2024b 이후) |
 | IncrementalRegressionKernel Fit | Fit incremental kernel regression model (R2024b 이후) |
 | IncrementalRegressionKernel Predict | Predict responses using incremental kernel regression model (R2024b 이후) |
 | IncrementalClassificationNaiveBayes Fit | Fit incremental naive Bayes classification model (R2025a 이후) |
 | IncrementalClassificationNaiveBayes Predict | Classify observations using incremental naive Bayes classification model (R2025a 이후) |
 | Detect Drift | 드리프트 감지기 상태와 드리프트 상태를 새 데이터로 업데이트하기 (R2024b 이후) |
 | Per Observation Loss | Per observation regression or classification error of incremental model (R2025a 이후) |
 | Update Metrics | Update performance metrics in incremental learning model given new data (R2023b 이후) |
 | KNN Search | searcher 객체를 사용하여 k-최근접이웃 찾기 (R2023b 이후) |
 | Isolation Forest Anomaly Detector | Find anomalies in data using isolation forest (R2026a 이후) |
 | One-Class SVM Anomaly Detector | Find anomalies in data using one-class support vector machine (SVM) (R2026a 이후) |
 | Scikit-learn Model Predict | Predict responses using pretrained Python scikit-learn model (R2024a 이후) |
 | Custom Python Model Predict | Predict responses using pretrained custom Python model (R2024a 이후) |
도움말 항목
분류
- Predict Class Labels Using ClassificationSVM Predict Block
This example shows how to use the ClassificationSVM Predict block for label prediction in Simulink®. - Predict Class Labels Using ClassificationTree Predict Block
Train a classification decision tree model using the Classification Learner app, and then use the ClassificationTree Predict block for label prediction. - Predict Class Labels Using ClassificationLinear Predict Block
This example shows how to use the ClassificationLinear Predict block for label prediction in Simulink®. (R2023a 이후) - Predict Class Labels Using ClassificationECOC Predict Block
Train an ECOC classification model, and then use the ClassificationECOC Predict block for label prediction. (R2023a 이후) - Predict Class Labels Using ClassificationEnsemble Predict Block
Train a classification ensemble model with optimal hyperparameters, and then use the ClassificationEnsemble Predict block for label prediction. - Predict Class Labels Using ClassificationNaiveBayes Predict Block
Train a naive Bayes classification model, and then use the ClassificationNaiveBayes Predict block for label prediction. (R2024a 이후) - Predict Class Labels Using ClassificationNeuralNetwork Predict Block
Train a neural network classification model, and then use the ClassificationNeuralNetwork Predict block for label prediction. - Predict Class Labels Using ClassificationKNN Predict Block
Train a nearest neighbor classification model, and then use the ClassificationKNN Predict block for label prediction. - Predict Class Labels Using ClassificationDiscriminant Predict Block
Train a discriminant analysis classification model, and then use the ClassificationDiscriminant Predict block for label prediction. (R2023b 이후) - Predict Class Labels Using ClassificationKernel Predict Block
Train a Gaussian kernel classification model, and then use the ClassificationKernel Predict block for label prediction. (R2024b 이후) - Predict Class Labels Using ClassificationXGBoost Predict Block
Use the ClassificationXGBoost Predict block for label prediction.
회귀
- Predict Responses Using RegressionSVM Predict Block
Train a support vector machine (SVM) regression model using the Regression Learner app, and then use the RegressionSVM Predict block for response prediction. - Predict Responses Using RegressionTree Predict Block
This example shows how to use the RegressionTree Predict block for response prediction in Simulink®. - Predict Responses Using RegressionLinear Predict Block
This example shows how to use the RegressionLinear Predict block for response prediction in Simulink®. (R2023a 이후) - Predict Responses Using RegressionEnsemble Predict Block
Train a regression ensemble model with optimal hyperparameters, and then use the RegressionEnsemble Predict block for response prediction. - Predict Responses Using RegressionNeuralNetwork Predict Block
Train a neural network regression model, and then use the RegressionNeuralNetwork Predict block for response prediction. - Predict Responses Using RegressionGP Predict Block
Train a Gaussian process (GP) regression model, and then use the RegressionGP Predict block for response prediction. - Predict Responses Using RegressionKernel Predict Block
This example shows how to use the RegressionKernel Predict block for response prediction in Simulink®. (R2024b 이후) - Predict Responses Using RegressionXGBoost Predict Block
Use the RegressionXGBoost Predict block for response prediction.
점진적 학습
- Perform Incremental Learning Using IncrementalClassificationLinear Fit and Predict Blocks
Perform incremental learning with the IncrementalClassificationLinear Fit block and predict labels with the IncrementalClassificationLinear Predict block. (R2023b 이후) - Perform Incremental Learning Using IncrementalRegressionLinear Fit and Predict Blocks
Perform incremental learning with the IncrementalRegressionLinear Fit block and predict responses with the IncrementalRegressionLinear Predict block. (R2023b 이후) - Perform Incremental Learning Using IncrementalClassificationECOC Fit and Predict Blocks
Perform incremental learning with the IncrementalClassificationECOC Fit block and predict labels with the IncrementalClassificationECOC Predict block. (R2024a 이후) - Perform Incremental Learning Using IncrementalClassificationKernel Fit and Predict Blocks
Perform incremental learning with the IncrementalClassificationKernel Fit block and predict labels with the IncrementalClassificationKernel Predict block. (R2024b 이후) - Perform Incremental Learning Using IncrementalRegressionKernel Fit and Predict Blocks
Perform incremental learning with the IncrementalRegressionKernel Fit block and predict responses with the IncrementalRegressionKernel Predict block. (R2024b 이후) - Perform Incremental Learning and Track Performance Metrics Using Update Metrics Block
Perform incremental learning and track performance metrics with the Update Metrics block. (R2023b 이후) - Monitor Drift Using Detect Drift Block
This example shows how to use the 드리프트 감지 block for monitoring drift in a data stream in Simulink®. (R2024b 이후) - In-Place Model Update of Offline Linear Model Using IncrementalClassificationLinear Predict Block
Perform in-place model update without regenerating deployed code. (R2025a 이후)
점진적 학습 템플릿
- Configure Simulink Template for Conditionally Enabled Incremental Linear Classification
Configure the Simulink Enabled Execution Incremental Learning template to perform incremental linear classification. (R2024a 이후) - Configure Simulink Template for Conditionally Enabled Incremental Linear Regression
Configure the Simulink Enabled Execution Incremental Learning template to perform incremental linear regression. (R2024a 이후) - Configure Simulink Template for Rate-Based Incremental Linear Classification
Configure the Simulink Rate-Based Incremental Learning template to perform incremental linear classification. (R2024a 이후) - Configure Simulink Template for Rate-Based Incremental Linear Regression
Configure the Simulink Rate-Based Incremental Learning template to perform incremental linear regression. (R2024a 이후) - Configure Simulink Template for Drift-Aware Incremental Learning
Configure the Drift-Aware Training for Incremental Learning template to perform drift-aware learning. (R2025a 이후)
군집 분석과 이상 감지
- Find Nearest Neighbors Using KNN Search Block
Train a nearest neighbor searcher model, and then use the KNN Search block for label prediction. (R2023b 이후)
Python 연동실행
- Predict Cluster Assignments Using Python Scikit-learn Model Predict Block
This example shows how to use the Scikit-learn Model Predict block for prediction in Simulink®. - Predict Responses Using Custom Python Model in Simulink
This example shows how to use the Custom Python Model Predict block for prediction in Simulink®.
Simulink로 학습기 앱 모델 내보내기
- Export Classification Model to Make Predictions in Simulink
After training a model in Classification Learner, export the model to Simulink. - Export Regression Model to Make Predictions in Simulink
After training a model in Regression Learner, export the model to Simulink.
코드 생성
- System Objects for Classification and Code Generation
Generate code from a System object™ for making predictions using a trained classification model, and use the System object in a Simulink model. - Predict Class Labels Using MATLAB Function Block
Generate code from a Simulink model that classifies data using an SVM model. - Predict Class Labels Using Stateflow
Generate code from a Stateflow® model that classifies data using a discriminant analysis classifier.
관련 정보
- Simulink를 사용한 딥러닝 (Deep Learning Toolbox)
추천 예제
Human Activity Recognition Simulink Model for Fixed-Point Deployment
Generate code from a classification Simulink model prepared for fixed-point deployment.
Human Activity Recognition Simulink Model for Smartphone Deployment
Generate code from a classification Simulink model prepared for deployment to a smartphone.
Identify Punch and Flex Hand Gestures Using Machine Learning Algorithm on Arduino Hardware
Use the Simulink® Support Package for Arduino® Hardware to identify punch and flex hand gestures using a machine learning algorithm. The example is deployed on an Arduino Nano 33 IoT hardware board that uses an onboard LSM6DS3 6DOF IMU sensor to identify the hand gestures. The output from the machine learning algorithm, after identifying whether a hand gesture is a punch or a flex, is transmitted to the serial port where 0 represents a punch and 1 represents a flex.
Compress Machine Learning Model for Memory-Limited Hardware
Reduce model size by feature selection, constrained Bayesian optimization, and parameter quantization.
