Help displaying a formatted matrix to the command window

Hi all, this seems like a simple question, but sprintf(...) doesn't do the trick. Is there an EASY way to display a matrix in the command window with a specified number of digits of precision?

EXAMPLE:

>> x = rand(3,2) - 0.5

x =

   -0.3810   -0.1596
   -0.0016    0.0853
    0.4597   -0.2762

>> format long >> x

x =

-0.381002318441623  -0.159614273333867
-0.001635948017857   0.085267750979777
 0.459743958516081  -0.276188060508863

How do I show x with, say, 6 digits of precision (as in below)??

-0.381002  -0.159614
-0.001635   0.085267
 0.459743  -0.276188

I'm sure one could write a code that figured out the size of x and then used sprintf to create the desired output. Does this code exist?

Answer 1

[Brenden Epps]

I have created a function called DISPN that answers this question.

http://www.mathworks.com/matlabcentral/fileexchange/37025

Answer 2

[Oleg Komarov]

Please read more carefully the documentation of sprintf(), it DOES the trick:

sprintf('%.6f %.6f\n',rand(3,2))

Edited

Now, if you wanna make the number of columns adjust automatically, then:

A = randn(3,9);
sprintf([repmat('%10.6f',1,size(A,2)) '\n'],A)

Unfortunately, there's some learning curve in format customization.

[Brenden Epps]

Hi Oleg, thanks for your quick reply. Yes, the whole point was to automate sprintf... What can be done about the issue of negative signs? sprintf fails in that case:

A = rand(3,9);

sprintf([repmat('%.6f',1,size(A,2) '\n'],A)

??? sprintf([repmat('%.6f',1,size(A,2) '\n'],A)

|

Error: Unexpected MATLAB expression.

>> sprintf([repmat('%.6f ',1,size(A,2)),'\n'],A)

ans =

0.751267 0.255095 0.505957 0.699077 0.890903 0.959291 0.547216 0.138624 0.149294

0.257508 0.840717 0.254282 0.814285 0.243525 0.929264 0.349984 0.196595 0.251084

0.616045 0.473289 0.351660 0.830829 0.585264 0.549724 0.917194 0.285839 0.757200

>> A = rand(3,9)-0.5;

>> sprintf([repmat('%.6f ',1,size(A,2)),'\n'],A)

ans =

0.253729 -0.119554 0.067822 -0.424146 -0.446050 0.030798 0.279167 0.434011 -0.370094

0.068824 -0.030609 -0.488098 -0.162877 -0.337818 0.294285 -0.188785 0.028533 -0.334351

0.101982 -0.237029 0.154079 0.189215 0.248152 -0.049458 -0.416179 -0.271023 0.413337

[Oleg Komarov]

I edited the second solution, now it's correct.

[Brenden Epps]

Thanks Oleg. I was hoping that the numbers would be aligned at the decimal place, as is the case when you display a matrix in the command window normally.

[Oleg Komarov]

A = randn(3,9);

sprintf([repmat('%10.6f',1,size(A,2)) '\n'],A)

Have you tried it?

You can tweak the %10 to give more space between the numbers.

[Jan]

+1: I still prefer the REPMAT approach. And if the minus character matters, I'd use the '+' specification and replace the '+' by spaces:

A = 1 - randn(3,9); fprintf(strrep(sprintf([repmat('%+10.6f',1,size(A,2)) '\n'],A), '+', ' '));

[Anton]

It seems that this one transposes my matrix in the process though

[Stephen23]

@Anton: yes, all of the above examples should transpose the array, like this:

fprintf(...,A.')

The fprintf help explains that it "...applies the formatSpec to all elements of arrays A1,... An in column order, and writes the data..." (emphasis added).

Answer 3

[Walter Roberson]

Try

num2str(x, 6)