분류 학습기 앱
대화형 방식으로 분류 모델 훈련, 검증 및 조정
이진 문제 또는 다중클래스 문제에 대해 분류 모델을 훈련시키고 검증하는 다양한 알고리즘 중에서 선택할 수 있습니다. 여러 모델을 훈련시킨 후 검증 오차를 나란히 비교한 다음 최적의 모델을 선택합니다. 어떤 알고리즘을 사용할지 결정하는 데 도움이 필요하다면 분류 학습기 앱에서 분류 모델을 훈련시키기 항목을 참조하십시오.
다음 플로우 차트는 분류 학습기 앱에서 분류 모델 또는 분류기를 훈련시키는 일반적인 워크플로를 보여줍니다.
분류 학습기에서 학습시킨 모델 중 하나를 사용하여 실험을 실행하려는 경우 해당 모델을 실험 관리자 앱으로 내보낼 수 있습니다. 자세한 내용은 Export Model from Classification Learner to Experiment Manager 항목을 참조하십시오.
회귀 모델을 훈련하고 검증하는 방법을 알아보려면 회귀 학습기 항목을 참조하십시오.
앱
도움말 항목
일반 워크플로
- Start a Classification Learner or Regression Learner Session
Start an app session by importing data from a file or the workspace, or by opening a saved app session. - Select Validation Scheme in Classification Learner or Regression Learner
Select a validation scheme to examine the predictive accuracy of models that you train. - 분류 학습기 앱에서 분류 모델을 훈련시키기
자동화된 훈련, 수동 훈련, 병렬 훈련 등 분류 모델을 훈련시키고 비교하고 향상시킬 수 있는 워크플로입니다. - Choose Classifier Options in Classification Learner
In Classification Learner, automatically train a selection of models, or compare and tune options in decision tree, discriminant analysis, logistic regression, naive Bayes, support vector machine, nearest neighbor, kernel approximation, ensemble, and neural network models. - Import Trained Model from Workspace into Classification Learner or Regression Learner
Import a trained model, including its training data, from the workspace at the start of a new session, or import a compatible trained model during the current session. (R2026a 이후) - Train Decision Trees Using Classification Learner App
Create and compare classification trees, and export trained models to make predictions for new data.
사용자 지정 워크플로
- 분류 학습기 앱을 사용한 특징 선택 및 특징 변환
분류 학습기에서 플롯 또는 특징 순위 지정 알고리즘을 사용하여 유용한 예측 변수를 식별하고, 포함할 특징을 선택하고, PCA를 사용하여 특징을 변환합니다. - Hyperparameter Optimization in Classification Learner App
Automatically tune hyperparameters of classification models by using hyperparameter optimization. - Train Classifier Using Hyperparameter Optimization in Classification Learner App
Train a classification support vector machine (SVM) model with optimized hyperparameters. - Misclassification Costs in Classification Learner App
Before training any classification models, specify the costs associated with misclassifying the observations of one class into another. - Train and Compare Classifiers Using Misclassification Costs in Classification Learner App
Create classifiers after specifying misclassification costs, and compare the accuracy and total misclassification cost of the models. - Edit Customizable Neural Network Using Network Editor in Classification Learner or Regression Learner
Edit a customizable neural network using the Network Editor, and then train the model and use training progress plots to check for overfitting. (R2026a 이후) - Build Condition Model for Industrial Machinery and Manufacturing Processes
Train a binary classification model using Classification Learner App to detect anomalies in sensor data collected from an industrial manufacturing machine. - Human Activity Recognition Simulink Model for Smartphone Deployment
Generate code from a classification Simulink® model prepared for deployment to a smartphone.
모델 성능 평가하기
- 분류 학습기의 분류기 성능 시각화 및 평가하기
모델 정확도 값을 비교하고, 클래스 예측값을 플로팅하여 결과를 시각화하고, 혼동행렬에서 클래스별 성능을 검사합니다. - Check Classifier Performance Using Test Data Set in Classification Learner App
Import a test data set into Classification Learner, and check the test metrics for the best-performing trained models. - Use Partial Dependence Plots to Interpret Classifiers Trained in Classification Learner App
Determine how features are used in trained classifiers by creating partial dependence plots. - Test Trained Models in Classification Learner or Regression Learner
Test trained models to assess performance in real-world scenarios with unseen data. - Explain Model Predictions for Classifiers Trained in Classification Learner App
To understand how trained classifiers use predictors to make predictions, use global and local interpretability tools, such as permutation importance plots, partial dependence plots, LIME values, and Shapley values.
모델, 분할, 데이터 세트, 플롯 내보내기
- Export Classification Model to Predict New Data
After training a model in Classification Learner, export the model to the workspace to make predictions on new data, and deploy the model to MATLAB® Compiler™. - Export Classification Model to MATLAB Coder to Generate C/C++ Code
After training a model in Classification Learner, export the model to MATLAB Coder™ to generate C/C++ code for prediction. - Generate MATLAB Code to Train Model with New Data
After training a model in Classification Learner, generate MATLAB code. - Generate Code at Command Line Using Model Exported from Machine Learning App
Train a classification model using the Classification Learner app, and generate C/C++ code for prediction at the MATLAB command line. - Code Generation for Binary GLM Logistic Regression Model Trained in Classification Learner
This example shows how to train a binary GLM logistic regression model using Classification Learner, and then generate C code that predicts labels using the exported classification model. - Export Classification Model to Make Predictions in Simulink
After training a model in Classification Learner, export the model to Simulink. - Export Classification Model for Deployment to MATLAB Production Server
After training a model in Classification Learner, export the model for deployment to MATLAB Production Server™. - Deploy Model Trained in Classification Learner to MATLAB Production Server
Train a model in Classification Learner and export it for deployment to MATLAB Production Server. - Export Partitions and Data Sets from Classification Learner or Regression Learner
In Classification Learner and Regression Learner, export validation partitions, test partitions, and data sets to the workspace. (R2026a 이후) - Export Plots in Classification Learner App
Export and customize plots created before and after training.
실험 관리자 워크플로
- Export Model from Classification Learner to Experiment Manager
Export a classification model to Experiment Manager to perform multiple experiments. - Tune Classification Model Using Experiment Manager
Use different training data sets, hyperparameters, and visualizations to tune an efficient linear classifier in Experiment Manager.
관련 정보
- MATLAB의 머신러닝
- 실험 관리하기 (Deep Learning Toolbox)
교육 관련 자료
생명과학 분야의 머신러닝
생물학적 사례를 기반으로 머신러닝의 기본 사항을 알아봅니다.
