Problems of convert decimal number to a string

Now I want to convert a input decimal number (maybe a input number, not known) to a string, I try to use num2str and string function, but there are some problems that I can not solve.

num = [110.00001, 10.00000000000000001,99.01]';
% I want to num become ["110.00001", "10.00000000000000001","99.01"]
num2str([110.00001, 10.00000000000000001,99.01])

By use num2str it will delete some decimal places, I don't know how to do it, holp someone would help me.

Answer 1

[madhan ravi]

You cannot do that.

[CARTER LYBBERT]

Nice

[Walter Roberson]

The answer is correct, but see https://www.mathworks.com/matlabcentral/answers/555997-problems-of-convert-decimal-number-to-a-string#comment_916237 for explanation.

Answer 2

[Ajay Kumar]

You can define precision property in num2str. For example:

num = [110.00001, 10.00000000000000001,99.01]';
your_ans = num2str(num,10);

[madhan ravi]

"I want to num become ["110.00001", "10.00000000000000001","99.01"]"

[zongxian]

The result is ['110.00001', ' 10', ' 99.01']

Not ["110.00001", "10.00000000000000001","99.01"]

[Ajay Kumar]

Numeric variable style long has fixed-decimal format with 15 digits after the decimal point for double values. To deal with more than 15 decimal places you need to have some high precision library such as symbolic math toolbox.

https://www.mathworks.com/help/releases/R2020a/matlab/ref/format.html#btiwmh5-1-style

https://www.mathworks.com/matlabcentral/answers/252213-how-do-i-set-matlab-to-read-an-input-with-more-than-15-decimal-places

[Walter Roberson]

double is not fixed-decimal format with 16 digits after the decimal point: double is floating point format with 53 effective bits, which happens to work out as between 15 and 16 relative digits. You can reliably get 15 relative digits, but you cannot reliably get 16 digits; it depends on the exact range of values.

[zongxian]

I have an idea, if we set the decimal number as num2str(10.0000000001, '%10.20f'), and then delete the zeros one by one (from the last one digits), until the non-zeros digit, now I have try it, it show '10.00000000010000000827', why the last digits are not zeros??

[Walter Roberson]

The exact value stored for 10.0000000001 is

10.0000000001000000082740370999090373516082763671875

Values down to 10.000000000000001 can be distinguished from 10.0

10.0000000000000017763568394002504646778106689453125

It is important to remember for this purpose that double() does not use decimal representation: it uses IEEE 754 Double Precision Binary representation, which is a 64 bit representation with 1 sign bit, 11 exponent bits, and 52 explicit fraction bits, plus one "hidden" fraction bit (it is hidden in the exponent in a way.)

Answer 3

[KSSV]

iwant = sprintf("0.12f 0.12f 0.12f",num) ;

Try with sprintf.

You can split each number into a cell using num2cell, and use cellfun with sprintf. Or run a loop for each num.

[zongxian]

This way need give the format of the number, but if we don't know the detail decimal number (i.e. '0.12f') will return wrong answer.