Dear all,
I have to round a number to a specified number of decimal points.Anybody can help me??
Thank you

Plus de réponses (2)

Walter Roberson
Walter Roberson le 18 Oct 2011

0 votes

That cannot be done exactly in binary floating point arithmetic. Just the same way that you cannot exactly represent 1/3 in a fixed number of decimal digits, 1/100 cannot be exactly represented in binary digits.
It is possible to display something with two decimal places, such as by using
disp(sprintf('%.2f', 343.945584))
but you cannot store 343.95 exactly in a floating point number: try
sprintf('%.99f\n', 343.95)
and notice that it is not exactly 343.95 that is stored.
If you do not mind that you can only numerically get a close approximation, then use:
round(N*100)/100

1 commentaire

Image Analyst
Image Analyst le 18 Oct 2011
Walter, as you know this is a frequently asked question, since you've repeated it bazillions of times. Yet it's not in the FAQ (that I could find anyway). I'm thinking it should be added to section 6 on Math/Algorithms. What do you think? You can add it, or I could if you want.

Connectez-vous pour commenter.

Pritesh Shah
Pritesh Shah le 18 Oct 2011

0 votes

Also, Please see Cell, fix and round function for the same. If you want decimal point fix,
Also, You can you format function.
Thank you

1 commentaire

Walter Roberson
Walter Roberson le 18 Oct 2011
I think you mean ceil rather than cell.
The "format" function cannot be used to get a precise number of decimal places, except for "format bank"

Connectez-vous pour commenter.

Community Treasure Hunt

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

Start Hunting!

Translated by