ParaMonte Fortran 2.0.0
Parallel Monte Carlo and Machine Learning Library
See the latest version documentation.
pm_arraySplit Module Reference

This module contains procedures and generic interfaces for splitting arrays of various types at the specified instances of occurrence of pattern. More...

Data Types

interface  setSplit
 Return the parts of the input array split at the requested occurrences of the input sep. More...
 

Variables

character(*, SK), parameter MODULE_NAME = "@pm_arraySplit"
 

Detailed Description

This module contains procedures and generic interfaces for splitting arrays of various types at the specified instances of occurrence of pattern.

Benchmarks:


Benchmark :: The runtime performance of setSplit for scalar vs. vector input sep argument.

1! Test the performance of split with `vector_sep` vs. `scalar_sep`.
2program benchmark
3
4 use iso_fortran_env, only: error_unit
5 use pm_container, only: cvi_type
6 use pm_kind, only: IK, LK, RK, SK
7 use pm_bench, only: bench_type
8
9 implicit none
10
11 integer(IK) :: i
12 integer(IK) :: isize
13 integer(IK) :: fileUnit
14 integer(IK) , parameter :: NSIZE = 15_IK
15 integer(IK) , parameter :: NBENCH = 2_IK
16 integer(IK) :: arraySize(NSIZE)
17 logical(LK) :: dummy = .true._LK
18 integer(IK) , allocatable :: array(:)
19 integer(IK) , parameter :: sep(1) = 0_IK
20 type(bench_type) :: bench(NBENCH)
21 type(cvi_type), allocatable :: ArraySplit(:)
22
23 bench(1) = bench_type(name = SK_"scalar_sep", exec = scalar_sep , overhead = setOverhead)
24 bench(2) = bench_type(name = SK_"vector_sep", exec = vector_sep , overhead = setOverhead)
25
26 arraySize = [( 2_IK**isize, isize = 1_IK, NSIZE )]
27
28 write(*,"(*(g0,:,' '))")
29 write(*,"(*(g0,:,' '))") "scalar_sep() vs. vector_sep()"
30 write(*,"(*(g0,:,' '))")
31
32 open(newunit = fileUnit, file = "main.out", status = "replace")
33
34 write(fileUnit, "(*(g0,:,','))") "arraySize", (bench(i)%name, i = 1, NBENCH)
35
36 loopOverArraySize: do isize = 1, NSIZE
37
38 write(*,"(*(g0,:,' '))") "Benchmarking with size", arraySize(isize)
39 allocate(array(arraySize(isize)))
40
41 do i = 1, NBENCH
42 bench(i)%timing = bench(i)%getTiming(minsec = 0.05_RK)
43 end do
44
45 deallocate(array)
46 write(fileUnit,"(*(g0,:,','))") arraySize(isize), (bench(i)%timing%mean, i = 1, NBENCH)
47
48 end do loopOverArraySize
49 write(*,"(*(g0,:,' '))") dummy
50 write(*,"(*(g0,:,' '))")
51
52 close(fileUnit)
53
54contains
55
56 !%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
57 ! procedure wrappers.
58 !%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
59
60 subroutine setOverhead()
61 call initialize()
62 call finalize()
63 end subroutine
64
65 subroutine initialize()
66 array(:) = 1_IK
67 end subroutine
68
69 subroutine finalize()
70 dummy = dummy .and. size(ArraySplit, kind = IK) == 1_IK
71 end subroutine
72
73 subroutine scalar_sep()
74 use pm_arraySplit, only: setSplit
75 call initialize()
76 call setSplit(ArraySplit, array, sep(1))
77 call finalize()
78 end subroutine
79
80 subroutine vector_sep()
81 block
82 use pm_arraySplit, only: setSplit
83 call initialize()
84 call setSplit(ArraySplit, array, sep)
85 call finalize()
86 end block
87 end subroutine
88
89end program benchmark
Return the parts of the input array split at the requested occurrences of the input sep.
Generate and return an object of type timing_type containing the benchmark timing information and sta...
Definition: pm_bench.F90:574
This module contains procedures and generic interfaces for splitting arrays of various types at the s...
This module contains abstract interfaces and types that facilitate benchmarking of different procedur...
Definition: pm_bench.F90:41
This module contains the derived types for generating allocatable containers of scalar,...
This module defines the relevant Fortran kind type-parameters frequently used in the ParaMonte librar...
Definition: pm_kind.F90:268
integer, parameter RK
The default real kind in the ParaMonte library: real64 in Fortran, c_double in C-Fortran Interoperati...
Definition: pm_kind.F90:543
integer, parameter LK
The default logical kind in the ParaMonte library: kind(.true.) in Fortran, kind(....
Definition: pm_kind.F90:541
integer, parameter IK
The default integer kind in the ParaMonte library: int32 in Fortran, c_int32_t in C-Fortran Interoper...
Definition: pm_kind.F90:540
integer, parameter SK
The default character kind in the ParaMonte library: kind("a") in Fortran, c_char in C-Fortran Intero...
Definition: pm_kind.F90:539
This is the class for creating benchmark and performance-profiling objects.
Definition: pm_bench.F90:386
This is the derived type for generating a container of a vector component of type integer of default ...

Example Unix compile command via Intel ifort compiler
1#!/usr/bin/env sh
2rm main.exe
3ifort -fpp -standard-semantics -O3 -Wl,-rpath,../../../lib -I../../../inc main.F90 ../../../lib/libparamonte* -o main.exe
4./main.exe

Example Windows Batch compile command via Intel ifort compiler
1del main.exe
2set PATH=..\..\..\lib;%PATH%
3ifort /fpp /standard-semantics /O3 /I:..\..\..\include main.F90 ..\..\..\lib\libparamonte*.lib /exe:main.exe
4main.exe

Example Unix / MinGW compile command via GNU gfortran compiler
1#!/usr/bin/env sh
2rm main.exe
3gfortran -cpp -ffree-line-length-none -O3 -Wl,-rpath,../../../lib -I../../../inc main.F90 ../../../lib/libparamonte* -o main.exe
4./main.exe

Postprocessing of the benchmark output
1#!/usr/bin/env python
2
3import matplotlib.pyplot as plt
4import pandas as pd
5import numpy as np
6
7fontsize = 14
8
9methods = ["scalar_sep", "vector_sep"]
10
11df = pd.read_csv("main.out")
12
13
16
17ax = plt.figure(figsize = 1.25 * np.array([6.4,4.6]), dpi = 200)
18ax = plt.subplot()
19
20for method in methods:
21 plt.plot( df["arraySize"].values
22 , df[method].values
23 , linewidth = 2
24 )
25
26plt.xticks(fontsize = fontsize)
27plt.yticks(fontsize = fontsize)
28ax.set_xlabel("Array Size", fontsize = fontsize)
29ax.set_ylabel("Runtime [ seconds ]", fontsize = fontsize)
30ax.set_title("Splitting array with sep(1) (scalar) vs. sep(1:1) (vector).\nLower is better.", fontsize = fontsize)
31ax.set_xscale("log")
32ax.set_yscale("log")
33plt.minorticks_on()
34plt.grid(visible = True, which = "both", axis = "both", color = "0.85", linestyle = "-")
35ax.tick_params(axis = "y", which = "minor")
36ax.tick_params(axis = "x", which = "minor")
37ax.legend ( methods
38 #, loc='center left'
39 #, bbox_to_anchor=(1, 0.5)
40 , fontsize = fontsize
41 )
42
43plt.tight_layout()
44plt.savefig("benchmark.scalarSep_vs_vectorSep.runtime.png")
45
46
49
50ax = plt.figure(figsize = 1.25 * np.array([6.4,4.6]), dpi = 200)
51ax = plt.subplot()
52
53plt.plot( df["arraySize"].values
54 , np.ones(len(df["arraySize"].values))
55 #, linestyle = "--"
56 #, color = "black"
57 , linewidth = 2
58 )
59plt.plot( df["arraySize"].values
60 , df["vector_sep"].values / df["scalar_sep"].values
61 , linewidth = 2
62 )
63
64plt.xticks(fontsize = fontsize)
65plt.yticks(fontsize = fontsize)
66ax.set_xlabel("Array Size", fontsize = fontsize)
67ax.set_ylabel("Runtime compared to scalar_sep()", fontsize = fontsize)
68ax.set_title("Runtime Ratio: split with sep(1:1) / split with sep(1).\nLower means faster. Lower than 1 means faster than scalar_sep.", fontsize = fontsize)
69ax.set_xscale("log")
70#ax.set_yscale("log")
71plt.minorticks_on()
72plt.grid(visible = True, which = "both", axis = "both", color = "0.85", linestyle = "-")
73ax.tick_params(axis = "y", which = "minor")
74ax.tick_params(axis = "x", which = "minor")
75ax.legend ( ["scalar_sep", "vector_sep"]
76 #, bbox_to_anchor = (1, 0.5)
77 #, loc = "center left"
78 , fontsize = fontsize
79 )
80
81plt.tight_layout()
82plt.savefig("benchmark.scalarSep_vs_vectorSep.runtime.ratio.png")

Visualization of the benchmark output

Benchmark moral
  1. The procedures under the generic interface setSplit take both scalar and vector sep arguments.
    As evidenced by the above benchmark, when the input sep is vector of length 1, it is much faster, up to 4X, to pass sep as a scalar instead of a whole array of length 1.
    Note that this benchmark is likely irrelevant to removing substrings from Fortran strings.
Test:
test_pm_arraySplit
Todo:
Normal Priority: A benchmark comparing the performance of output index array vs. output jagged array would be informative here.


Final Remarks


If you believe this algorithm or its documentation can be improved, we appreciate your contribution and help to edit this page's documentation and source file on GitHub.
For details on the naming abbreviations, see this page.
For details on the naming conventions, see this page.
This software is distributed under the MIT license with additional terms outlined below.

  1. If you use any parts or concepts from this library to any extent, please acknowledge the usage by citing the relevant publications of the ParaMonte library.
  2. If you regenerate any parts/ideas from this library in a programming environment other than those currently supported by this ParaMonte library (i.e., other than C, C++, Fortran, MATLAB, Python, R), please also ask the end users to cite this original ParaMonte library.

This software is available to the public under a highly permissive license.
Help us justify its continued development and maintenance by acknowledging its benefit to society, distributing it, and contributing to it.

Author:
Fatemeh Bagheri, Wednesday 12:20 AM, October 13, 2021, Dallas, TX

Variable Documentation

◆ MODULE_NAME

character(*, SK), parameter pm_arraySplit::MODULE_NAME = "@pm_arraySplit"

Definition at line 54 of file pm_arraySplit.F90.