1. There are three kinds of arithmetic precision in Matlab, they are numerical algorithm, symbolic algorithm and controllable precision algorithm. They are introduced as follows:
(1) Numerical algorithm takes each number as 16 bits, and the calculation is performed by floating point operation, which is the fastest algorithm.
(2) Symbolic algorithm converts each number into a symbolic quantity, and the operation is carried out according to rational quantity calculation.
(3) The controllable precision algorithm is between the above two algorithms, and it can make the calculation under controllable precision.
2. The data display format of Matlab, the list is as follows:
Table Matlab data display format command
Order |
significance |
Example ( ) |
format short |
Short format mode, display 5-digit fixed-point decimal number |
3.1416 |
format long |
Long format, display 15-bit fixed-point decimal number |
3.141592653589793 |
format short e |
Optimized short format display, 5 digits plus exponent |
3.1416e+000 |
format long e |
Optimal format, 15 digits plus exponent |
3.141592653589793e+000 |
format short g |
5-bit fixed-point or floating-point format |
3.1416 |
format long g |
For double precision, display 15-bit fixed-point or floating-point format, for single-precision, display 7-bit fixed-point or floating-point format |
3.14159265358979 |
format short eng |
At least 5 digits plus 3 digits for exponent |
3.1416e+000 |
format long eng |
16 bits plus at least 3 bits of exponent |
3.1415926358979e+000 |
format hex |
hex format |
400921fb54442d18 |
format bank |
Bank format. Fixed format by yuan, angle, minute (with two decimal places) |
3.14 |
format + |
+ format, with +, - and space for positive, negative and zero elements in |
+ |
format |
The default is the default short format, the same as format short |
3.1416 |
format rat |
Fractional format form. Approximate display data with rational numbers |
355/113 |
format loose |
Loose format. Blank lines between data |
|
format compact |
Compact format. No blank lines between data |
|
vpa(date,n) |
Display data date with n significant digits |
vpa(pi,5)= 3.1416 |
The format does not affect how matlab calculates and stores the value of the variable. Calculations on floating point variables, i.e. single or double precision, are performed with the appropriate floating point precision, regardless of how the variable is displayed. Use integer for integer variables Type data. Integer variables are always displayed with appropriate data bits according to different classes.
3. Matlab special variables
ans: reaction to recent input
computer: the current computer type
eps: floating point precision
flops: counts the number of floating-point operations, no longer commonly used
i: imaginary unit
inf: infinity
inputname: input parameter name
j: imaginary unit
nan: non-numeric
nargin: the number of input parameters
nargout: number of output arguments (user-defined function)
pi: pi
realmax: maximum number of positive floating points
realmin: minimum number of positive floating points
varargin
varargout: returns the number of parameters (matlab function)
cputime: CPU working time