Why two variables loading the same .mat file are not equal ?
4 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Florian Bidaud
el 17 de Feb. de 2022
Comentada: Rik
el 17 de Feb. de 2022
Hi,
I found something weird today and couldn't find an explanation on the forum.
When I load the same matlab file twice, and try to check if they are equal, the output is False.
In my situation, a and b are structures with large arrays inside, and I manually checked that they are actually equal.
Where does this come from ? Is there another way to quickly check the equality ?
Thanks !
a = load('myFile.mat')
b = load('myFile.mat')
isequal(a,b)
ans =
logical
0
3 comentarios
Image Analyst
el 17 de Feb. de 2022
Can you at least show us a screenshot of you doing those 3 lines of code? And have both the code in the editor window, and the command window below in the picture. We need to make sure that they are both exactly the same filename, and not different somehow (like one ends in a space or something).
Respuesta aceptada
Rik
el 17 de Feb. de 2022
Editada: Rik
el 17 de Feb. de 2022
I guess boldly:
There might be a NaN value hidden somewhere in your data:
isequal(NaN,NaN)
s1=struct('x',NaN);s2=s1;
isequal(s1,s2)
isequal(ComputeNonCryptHash(s1),ComputeNonCryptHash(s2))
Más respuestas (2)
Walter Roberson
el 17 de Feb. de 2022
You can get results that are not isequal() when you load data that does not serialize exactly.
Consider for example if you save a graphics object, then when you load() it is going to create two different versions of the graphics object.
0 comentarios
Florian Bidaud
el 17 de Feb. de 2022
3 comentarios
Rik
el 17 de Feb. de 2022
And Walter added another explanation: if the mat file contains a graphics object the second load might create a new object, instead of remembering that it has already that object in memory.
Ver también
Categorías
Más información sobre Data Type Identification en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!

