problem with 'sum' command

hello if i have this code: x=[10,1000] a=sum(x) i get after the sum that a=[1,1000] so i need to find the inverse command for 'sum' to create the original [10,1000] matrix how does i do that? maybe with 'fft'? thanks!

Answer 1

[the cyclist]

You mean that the size of x is 10x10000, right? Not that x is the vector [10,10000], which is what you wrote.

There is no unique inverse to the sum() command. Think about it. If I tell you that 2+2 equals 4, there is no way to for you to undo that, and obtain 2 and 2 from 4. It could have been 1 and 3, or 0 and 4, or -1.1 and 5.1, etc.

[John D'Errico]

I recall hearing the next release of matlab is supposed to have the unsum function in it, as part of the clairvoyance toolbox. They will naturally move the function magic.m into the clairvoyance toolbox at that time.

Answer 2

[itsik]

u right my friend bu im asking that, maybe there is a way to sample the sum(x) and mak an fft to open it?

[the cyclist]

I think you would need to provide a lot more detail about what you want as output from this function, to get some help.

[Walter Roberson]

I would go further and say NO outright, that there is no way to proceed given only the single sum.

If there was at least a sum along the rows and a sum along the columns then there might be something that could be done.