Does anyone know if cftool can handle fits of data with more than 3 dimensions? i.e., hyperdimensional surface fits? Thanks in advance.

 採用された回答

Matt J
Matt J 2021 年 1 月 14 日
編集済み: Matt J 2021 年 1 月 14 日

1 投票

3 件のコメント

Clay Fulcher
Clay Fulcher 2021 年 1 月 14 日
Thanks Matt for your fast reply. lsqcurvefit looks to be set up like fmincon, which I am familiar with. To use lsqcurvefit, would I need to define the spline fitting algorithm in the objective function myself? Or is there a pre-packaged spline fit similar to "thin-plate spline" used in cftool that I can use with lsqcurvefit?
Matt J
Matt J 2021 年 1 月 14 日
編集済み: Matt J 2021 年 1 月 14 日
You haven't mentioned anything about splines in your original post, so I don't know where it fits into anything...
You need to supply a function which generates the hypersurface for any guess of the parameters.
Clay Fulcher
Clay Fulcher 2021 年 1 月 14 日
Ok, thanks. That's what it seemed to me. Yes, this data does not fit any kind of closed-form model that I can think of very well. I've been using the cftool interpolation option for this data, and the "thin-plate" spline does a decent job. One day I'll try to write a hyperdimensional spline fitter, but I'm in kind of a hurry. Thanks for your help.

サインインしてコメントする。

その他の回答 (0 件)

カテゴリ

ヘルプ センター および File ExchangeGet Started with Curve Fitting Toolbox についてさらに検索

製品

リリース

R2020a

質問済み:

2021 年 1 月 14 日

コメント済み:

2021 年 1 月 14 日

Community Treasure Hunt

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

Start Hunting!

Translated by