stats

Definition

Provides methods to compute mean, standard deviation and variance of a given sample.

Examples

use benchmark_statistics
 
type(stats) :: s
integer     :: i
real(r8)    :: array(10)
 
do i = 1, 10
     array(i) = real(i, r8)
end do
 
call s%compute(array)
write(*, '(A)') str(s%mean, '(f12.3)') // ' ms'//' +/- '//str(s%stddev, '(f12.3)') // ' ms'

Remarks

The statistics computed at the moment are limited to mean, min, max and standard deviations. The implementation is quite naive and serve only the purpose of the current library.

Constructors

Initializes a new instance of the stats class

stats(integer, real(r8), real(r8), real(r8))

type(stats) :: s 

Examples

use benchmark_statistics
 
type(stats) :: s

Remarks

Definition at line 48 of file Statistics.f90.

---

The documentation for this type was generated from the following file:

• Statistics.f90

Variables
real(r8), public	mean = 0.0_r8
	Mean value of the sample.
real(r8), public	min = 0.0_r8
	Minimal value of the sample.
real(r8), public	max = 0.0_r8
	Maximal value of the sample.
real(r8), public	stddev = 0.0_r8
	Standard deviation of the sample.
real(r8), public	variance = 0.0_r8
	Variance of the sample.

Private Attributes
integer	n
	Sample size.

Methods

compute()

procedure, pass(this) compute	(	class(stats), intent(inout)	this,
		real(r8), dimension(:), intent(in)	y )

Compute mean, standard deviation and variance of a Type bound to the method.

Parameters

[in,out]	this	bound argument
[in]	y	Sample array

Examples

type(stats) :: s
integer :: i
real(r8) :: array(10)
 
call random_number(array)
call s%compute(array)

Note

At the moment it only works for real(r8) arrays

Remarks

Definition at line 57 of file Statistics.f90.