accélérer les algorithm

**yan** · 26/11/2007, 16h09

Bonjour,
constatant que les algo utilise souvent la fonction swap. J'ai essayer de jouer avec pour optimiser un sort.
Voici le code :

Code :

Sélectionner tout - Visualiser dans une fenêtre à part

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
 
#include <algorithm>
#include <iostream>
#include <vector>
#include <string>
 
//class qui contient la fonction swap pour éviter des recopie de memoire
class A 
{
public :
	A(int a,std::string s) :myString(s),nb(a){ };
 
	const std::string &GetString() {return myString;}
	bool operator<(const A& a) {return nb<a.nb;}
	A&  operator=(const A& a) 
		{
		static int nbegale(0);
		std::cout<<"= : "<<++nbegale<<std::endl;
		nb<a.nb;
		myString=a.myString;
		return (*this);
		}
	void swap(A& a) 
		{
			static int nbswap(0);
			std::cout<<"swap : "<<++nbswap<<std::endl;
			std::swap(nb,a.nb);
			myString.swap(a.myString);
		};
private:
	std::string myString;
	int nb;
 
};
//specialisation de la fonction swap pour la class A
template<>
inline void std::swap<A> ( A& a, A& b )
{
	a.swap(b);
};
 
int main(int argc, char** argv)
{
	//mon vecteur a trier
	std::vector<A> vect;
       //on remplie de n'importe quoi
      //le but etant quil y en ai assez pour que swap soit appelé
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
 
     //on fait le trie
	std::sort(vect.begin(),vect.end());
return 0;
}

En prenant l'hypothèse qu'un égale dure 1 second (du au reallocation et recopie de memoire), et swap .5s (echange de la mémoire, donc peut de recopie)

Les résultats que j'obtiens sous visual 20005 sont :
- sans le swap : 234 operation egale = 234s
- avec le swap : 74 operation swap et 72 operation egale = 109 s

Ce qui n'est pas négligable.
Maintenant je me demandé si :
1- ce que j'ai fait est tout à fait correcte
2- il n'y as pas d'autre manière de faire
3- on est obligé de spécifier std::swap

[EDIT]
4- existe t'il d'autre astuce comme celle la???
merci

**yan** · 26/11/2007, 18h54

voici une rversion qui marche avec GCC

Code C++ :

Sélectionner tout - Visualiser dans une fenêtre à part

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
 
#include <algorithm>
#include <iostream>
#include <vector>
#include <string>
 
//class qui contient la fonction swap pour éviter des recopie de memoire
class A
{
public :
	A(int a,std::string s) :myString(s),nb(a){ };
 
	const std::string &GetString() {return myString;}
	bool operator<(const A& a)const {return nb<a.nb;}
	A&  operator=(const A& a)
		{
		static int nbegale(0);
		std::cout<<"= : "<<++nbegale<<std::endl;
		nb=a.nb;
		myString=a.myString;
		return (*this);
		}
	void swap(A& a)
		{
			static int nbswap(0);
			std::cout<<"swap : "<<++nbswap<<std::endl;
			std::swap(nb,a.nb);
			myString.swap(a.myString);
		};
private:
	std::string myString;
	int nb;
 
};
//specialisation de la fonction swap pour la class A
namespace std
{
template<>
inline void swap<A> ( A& a, A& b )
{
	a.swap(b);
};
}
 
int main(int argc, char** argv)
{
	//mon vecteur a trier
	std::vector<A> vect;
       //on remplie de n'importe quoi
      //le but etant quil y en ai assez pour que swap soit appelé
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(2,"deux"));
	vect.push_back(A(15,"quinze"));
	vect.push_back(A(10,"dix"));
	vect.push_back(A(5,"cinq"));
	vect.push_back(A(1,"un"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
	vect.push_back(A(30,"trente"));
 
     //on fait le trie
	std::sort(vect.begin(),vect.end());
	return 0;
}

Les résultats que j'obtiens avec GCC sous ubuntu sont :
- sans le swap : 361 operation egale = 361s
- avec le swap : 88 operation swap et 185 operation egale = 229 s

**yan** · 26/11/2007, 18h58

Au faite, pour ceux qui veule tester,pour enlever le swap, il suffit de commenter la definition de std::swap

**yan** · 27/11/2007, 22h51

personne n'as d'idée???

**yan** · 27/11/2007, 23h08

Bon,
j'ai trouvé un truc pour ceux que cela intéresse :
http://www.tantalon.com/pete/cppopt/...l.htm#Swapping

**r0d** · 08/03/2008, 11h08

Il serait intéressant de faire des tests sur des gros conteneurs bien remplis. Et chronométrer, avec différentes méthodes, sur différentes plateformes, et avec une machine dédiée (je veux dire un machine où aucune appli ne tourne, pas même les démons/services réseau etc.)

accélérer les algorithm

SL & STL C++

Vue hybride

Discussions similaires

Partager

Partager