Is there any way to determne the number of support vectors while using classifier app?

1 回表示 (過去 30 日間)
krishna Chauhan
krishna Chauhan 2022 年 11 月 23 日
回答済み: Hornett 2024 年 9 月 25 日
I am using MATLAB classifer app and in that particularly classifier app. I need to compare the time complexity of SVM with corresponding ANN? In SVM the testing time complexity is O(kn), where k=number of support vectors and n=datasamples. For this I need k or number of support vectors.
Can someone guide on this please.
Thanks

回答 (1 件)

Hornett
Hornett 2024 年 9 月 25 日
Hi Krishna,
You can use the supportVectors propertt from the classificationSVM class to get the number of support vectors.
"Support vectors in the trained classifier, specified as an s-by-p numeric matrix. s is the number of support vectors in the trained classifier, sum(Mdl.IsSupportVector), and p is the number of predictor variables in the predictor data."
Hope it helps!

カテゴリ

Help Center および File ExchangeImage Data Workflows についてさらに検索

タグ

製品


リリース

R2021a

Community Treasure Hunt

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

Start Hunting!

Translated by