Issues with CV function in lasso

2019 5 23
0 个回答
5 次查看（30 天）

0 个投票

Hi there,
I am trying to run lasso on a small data set of metabolites, and keep getting an error when it comes to the cross validation.
Here is my code:
x=data;
y=phase;
[B,STATS]=lasso(x,y,'CV',10);
axTrace=lassoPlot(B,STATS);
axCV=lassoPlot(B,STATS,'PlotType','CV');
minMSEModel=STATS.PredictorNames(B(:,STATS.IndexMinMSE)~=0)
sparseModel=STATS.PredictorNames(B(:,STATS.Index1SE)~=0)
And here is my error message:
Warning: This format of input to CROSSVAL may be removed in the future. See 'help crossval'.
> In crossval (line 313)
In lasso (line 517)
In Lasso (line 3)
Undefined operator ':' for input arguments of type 'cvpartition'.
Error in crossval (line 405)
ncompvector = 1:ncomp;
Error in lasso (line 517)
cvMSE = crossval(cvfun,[weights(:) X],Y, ...
Error in Lasso (line 3)
[B,STATS]=lasso(x,y,'CV',10);
This how I have normally run lasso in the past, and appears to also be what is suggested online. Please help!

请先登录，再回答此问题。

回答（0 个）

类别

在 帮助中心 和 File Exchange 中查找有关 Gaussian Process Regression 的更多信息

产品

标签

Community Treasure Hunt

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

Start Hunting!

Translated by