Take the following example
x = single(-(-2:2).^2);
[xm,ix]=max(x);
whos
Name Size Bytes Class Attributes cmdout 1x33 66 char ix 1x1 8 double x 1x5 20 single xm 1x1 4 single
I think it strange that variable ix is of type double, and not integer, which would be the most logical alternative.
Anyway, I am using this kind of line for a code generation case, where double data type is banned.
Is there any way I can force max function to return an integer value for argmax (variable ix in code above)?

1 comentario

Steven Lord
Steven Lord el 13 de Mayo de 2022
I think it strange that variable ix is of type double, and not integer, which would be the most logical alternative.
I suspect that the two-output form of max predates the introduction of the integer data types to MATLAB.

Iniciar sesión para comentar.

 Respuesta aceptada

Matt J
Matt J el 13 de Mayo de 2022

0 votos

No, but can't you just cast it to an integer type post-facto? E.g.,
ix=cast(ix,'uint32');

1 comentario

Mats
Mats el 13 de Mayo de 2022
Hi. I did not even try this trick because I though that you cannot change the datatype of a variable when using code generation (which is generally true). However, becuase code generator is quite smart your suggestion works nicely.
Thank you Matt!

Iniciar sesión para comentar.

Más respuestas (0)

Categorías

Más información sobre Deployment, Integration, and Supported Hardware en Centro de ayuda y File Exchange.

Productos

Versión

R2019b

Preguntada:

el 13 de Mayo de 2022

Comentada:

el 13 de Mayo de 2022

Community Treasure Hunt

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

Start Hunting!

Translated by