how convert the vector to matrix ? unknown number of columns

Is it possible to convert the vector to matrix .unknown number of columns and the number of row is known. Where is the row filled with zeros? Or with holes?

A=[1 2 3 4 5 6 7 8 9]

number of row=3

A=[1 2 3 4
      5 6 7 8
      9 0 0 0]

[Matt J]

The task is ambiguous without additional specifications on how the number of columns is to be chosen. For example, with the input,

A=[1 2 3 4 5 6 7 8 9 10 11 12 13 14 15];
nrows=3;

there are at least three possible 3-row arrangements that would be valid:

B=[1 2 3 4 5
   6 7 8 9 10
   11 12 13 14 15]

B=[1 2 3 4 5 6
   7 8 9 10 11 12
   13 14 15 0 0 0]

B=[1 2 3 4 5 6 7
   8 9 10 11 12 13 14
   15 0 0 0 0 0 0]

Answer 1

[dpb]

A=1:9;
nRow-3;

>> A=reshape(A,3,[])
A =
     1     4     7
     2     5     8
     3     6     9
>> 

The posted in the Q? text above forced four columns, not three (3) rows...

That, of course, is also doable with a little more effort...

A=1:9;
nCol=4;
nZ=ceil(numel(A)/4);


>> reshape([A zeros(1,nCol*nZ-numel(A))],nCol,[]).'
ans =
     1     2     3     4
     5     6     7     8
     9     0     0     0
>>

[reta jon]

think you sir , if i have big vector and i want convert the vector to matrix ,unknown number of columns and the number of row is known.for exmple number of row 128 ,what i do ??

[dpb]

Size makes no difference to the logic -- to reshape() a vector to an NxM array the proudct N*M must be exactly the final number of elements in the vector. So, if they don't match; have to either

1. Truncate the vector to first even multiple of M columns such that Nprime*M = numel(V) or
2. Augment the vector V with the needed number of values to make N*M == numel(V)+numel(Vadd)

The above code logic shows how to do the latter for any V and choice of N or M; just the orientation controls which is chosen to be the desired fixed number -- rows or columns.

[Matt J]

if i have big vector and i want convert the vector to matrix ,unknown number of columns and the number of row is known.for exmple number of row 128 ,what i do ??

@reta jon, as has been pointed out to you in comments above, you must supply the desired number of columns. There is no unique number of columns that can be deduced given only the number of rows.

[dpb]

"There is no unique number of columns that can be deduced given only the number of rows." @Matt J

Well, one can always force N rows and let the columns fall where they may, but only by either truncating or augmenting such that the end result has same number of total elements and is divisible by N.

[Matt J]

@dpb yes, but augment to what? There isn't a unique choice.For the following input, for example,

A=[1 2 3 4 5 6 7 8 9 10 11 12 13 14 15]

there are at least three possible solutions with 3 rows:

B=[1 2 3 4 5
   6 7 8 9 10
   11 12 13 14 15]

B=[1 2 3 4 5 6
   7 8 9 10 11 12
   13 14 15 0 0 0]

B=[1 2 3 4 5 6 7
   8 9 10 11 12 13 14
   15 0 0 0 0 0 0]

[Image Analyst]

@Matt J, yes there are multiple possible answers. But since he did not actually ask how to construct any of them, and all he asked was "Is it possible to convert the vector to matrix .unknown number of columns and the number of row is known?" the answer is simply "Yes".

Answer 2

[Matt J]

No, the solution is ill-defined. Another result with the same number of rows is:

A=[1 2 3
   4 5 6 
   7 8 9]