MATLAB Answers

fitting 2 variable function to (x-1) form

1 ビュー (過去 30 日間)
Benjamin
Benjamin 2019 年 3 月 28 日
コメント済み: Benjamin 2019 年 3 月 29 日
I have this fit currently which works great:
fitobject = fit([r,eta/eta_c],H,ft,'problem',knownVals)
However, because of the nature of the data, it would make more sense if r was actually in the form of .
When I change the code to this:
fitobject = fit([(r-1),eta/eta_c],H,ft,'problem',knownVals)
My fit is completey messed up, and it looks nothing like the data.
My question is, am I employing the fit correctly?
The form should look like: Cij * (r-1)^i *(eta/eta_c)^j.
I wonder if I missed something in the fitting procedure?
  1 件のコメント
Walter Roberson
Walter Roberson 2019 年 3 月 28 日
Possibly it would help to use 'Upper' and 'Lower' (bounds) or 'StartPoint' options.

サインインしてコメントする。

採用された回答

Catalytic
Catalytic 2019 年 3 月 29 日
編集済み: Catalytic 2019 年 3 月 29 日
My fit is completey messed up, and it looks nothing like the data.
It shouldn't look like the r and eta data, because that's not the data you fed to the fit, but it should look a lot like r-1 and eta/eta_c.
If I were you, I would just re-define all the r and eta data and forget about it instead of repeatedly having to remember to transform them everywhere they are used in the code. So, I would have
r=r-1;
eta=eta/eta_c;
fitobject = fit([r,eta],H,ft,'problem',knownVals);
plot(fitobject,[r,eta],z)
  1 件のコメント
Benjamin
Benjamin 2019 年 3 月 29 日
Thanks for catching that mistake!!

サインインしてコメントする。

その他の回答 (0 件)

タグ

Community Treasure Hunt

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

Start Hunting!

Translated by