Bayesian Optimization? Can't find anything wrong.

1 次查看（过去 30 天）
yufan ji
yufan ji 2020-7-26
编辑： yufan ji 2020-7-26
% https://www.mathworks.com/help/stats/bayesopt.html
% I followed example in the first line and don't know how to change it right. Thanks a lot!
clear;clc;
load ionosphere
dataX = X;
dataY = Y;
c = cvpartition(351,'KFold',5);
kernel = optimizableVariable('kernel',{'gaussian','polynomial'},'Type','categorical');
kernelScale = optimizableVariable('kernelScale',[1, 30]);
polyOrder = optimizableVariable('polyOrder',[2, 3],'Type','integer');
vars = [kernel,kernelScale,polyOrder];
fun = @(x)kfoldLoss(fitcsvm(dataX,dataY,'CVPartition',c,'KernelFunction',string(x.kernel),'KernelScale',x.kernelScale,'PolynomialOrder',x.polyOrder));
results = bayesopt(fun,vars,'Verbose',1,'AcquisitionFunctionName','expected-improvement-plus');
%% The output
|====================================================================================================================|
| Iter | Eval | Objective | Objective | BestSoFar | BestSoFar | kernel | kernelScale | polyOrder |
| | result | | runtime | (observed) | (estim.) | | | |
|====================================================================================================================|
| 1 | Best | 0.11966 | 0.067042 | 0.11966 | 0.11966 | polynomial | 7.1654 | 3 |
错误使用 classreg.learning.partition.PartitionedModel/checkFoldArgs (line 327)
Indices of folds must be a vector with numbers between 1 and 0.
出错 classreg.learning.partition.ClassificationPartitionedModel/kfoldLoss (line 310)
[mode,folds,extraArgs] = checkFoldArgs(this,varargin{:});
出错
BO_SVM>@(x)kfoldLoss(fitcsvm(dataX,dataY,'CVPartition',c,'KernelFunction',string(x.kernel),'KernelScale',x.kernelScale,'PolynomialOrder',x.polyOrder))
出错 BayesianOptimization/callObjNormally (line 2553)
Objective = this.ObjectiveFcn(conditionalizeX(this, X));
出错 BayesianOptimization/callObjFcn (line 467)
= callObjNormally(this, X);
出错 BayesianOptimization/runSerial (line 1989)
ObjectiveFcnObjectiveEvaluationTime, ObjectiveNargout] = callObjFcn(this, this.XNext);
出错 BayesianOptimization/run (line 1941)
this = runSerial(this);
出错 BayesianOptimization (line 457)
this = run(this);
出错 bayesopt (line 323)
Results = BayesianOptimization(Options);
出错 BO_SVM (line 13)
results = bayesopt(fun,vars,'Verbose',1,'AcquisitionFunctionName','expected-improvement-plus');

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Classification Ensembles 的更多信息

标签

产品


版本

R2018a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by