Vad är Hamiltons cykel?
Hamiltonsk cykel eller krets i en graf G är en cykel som besöker varje hörn av G exakt en gång och återgår till startpunkten.
- Om grafen innehåller en Hamiltonsk cykel kallas den Hamiltonsk graf annars är det icke-hamiltonska .
- Att hitta en Hamiltonsk cykel i en graf är välkänt NP-komplett problem , vilket innebär att det inte finns någon känd effektiv algoritm för att lösa det för alla typer av grafer. Det kan dock lösas för små eller specifika typer av grafer.
Hamiltons cykelproblem har praktiska tillämpningar inom olika områden, som t.ex logistik, nätverksdesign och datavetenskap .
Vad är Hamiltonian Path?
Hamiltonska vägen i en graf G är en väg som besöker varje hörn av G exakt en gång och Hamiltonska vägen behöver inte återgå till startpunkten. Det är en öppen väg.
- Liknar Hamiltons cykel problem, att hitta en Hamiltonska vägen i en allmän graf är också NP-komplett och kan vara utmanande. Men det är ofta ett enklare problem än att hitta en Hamiltonsk cykel.
- Hamiltonian Paths har applikationer inom olika områden, som t.ex hitta optimala vägar inom transportnätverk, kretsdesign och grafteoretisk forskning .
Problembeskrivning: Givet en oriktad graf är uppgiften att avgöra om grafen innehåller en Hamiltonsk cykel eller inte. Om den innehåller, skrivs sökvägen ut.
Exempel:
Rekommenderas: Vänligen lös det ÖVA först innan du går vidare till lösningen.Inmatning: graf[][] = {{0, 1, 0, 1, 0},{1, 0, 1, 1, 1},{0, 1, 0, 0, 1},{1, 1, 0, 0, 1},{0, 1, 1, 1, 0}}
Inmatningsdiagram[][]
typskrift för varjeProduktion: {0, 1, 2, 4, 3, 0}.
Inmatning: graf[][] = {{0, 1, 0, 1, 0},{1, 0, 1, 1, 1},{0, 1, 0, 0, 1},{1, 1, 0, 0, 0},{0, 1, 1, 0, 0}}
Inmatningsdiagram[][]
Produktion: Lösningen finns inte
Naiv algoritm : Det här problemet kan lösas med följande idé:
Generera alla möjliga konfigurationer av hörn och skriv ut en konfiguration som uppfyller de givna begränsningarna. Det kommer att finnas n! (n faktoriella) konfigurationer. Så den övergripande tidskomplexiteten för detta tillvägagångssätt kommer att vara PÅ!).
Hamiltonian Cycle med hjälp av Backtracking-algoritm :
Skapa en tom banmatris och lägg till vertex 0 till det. Lägg till andra hörn, med början från hörnet 1 . Innan du lägger till en vertex, kontrollera om den ligger intill den tidigare tillagda vertexen och inte redan har lagts till. Om vi hittar en sådan vertex lägger vi till vertexen som en del av lösningen. Om vi inte hittar en vertex så återvänder vi falsk .
Illustrationer:
Låt oss ta reda på Hamiltons cykel för följande graf:
- Börja med noden 0 .
- Använd DFS för att hitta den Hamiltonska vägen.
- När basfallet når (dvs. totalt antal genomkörda noder == V (totalt vertex) ):
- Kontrollera om den aktuella noden är granne till startnoden.
- Som nod 2 och nod 0 är inte grannar med varandra så återvänd från det.
Börjar från startnod 0 anropar DFS
- Eftersom cykeln inte hittas i väg {0, 3, 1, 4, 2}. Så, återvänd från nod 2, nod 4.
- Utforska nu ett annat alternativ för nod 1 (dvs nod 2)
- När den träffar bastillståndet igen, kontrollera för Hamiltonsk cykel
- Eftersom nod 4 inte är granne till nod 0, återfinns inte cykeln och återgår sedan.
- Retur från nod 4, nod 2, nod 1.
- Utforska nu andra alternativ för nod 3.
Hamiltons cykel
- På Hamiltons väg {0,3,4,2,1,0} vi får cykel eftersom nod 1 är granne till nod 0.
- Så skriv ut den här cykelvägen.
- Detta är vår Hamiltonska cykel.
Nedan är Backtracking-implementationen för att hitta Hamiltonian Cycle:
C++ /* C++ program for solution of Hamiltonian Cycle problem using backtracking */ #include using namespace std; // Number of vertices in the graph #define V 5 void printSolution(int path[]); /* A utility function to check if the vertex v can be added at index 'pos' in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ bool isSafe(int v, bool graph[V][V], int path[], int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph [path[pos - 1]][ v ] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ bool hamCycleUtil(bool graph[V][V], int path[], int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the // last included vertex to the first vertex if (graph[path[pos - 1]][path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate // in Hamiltonian Cycle. We don't try for 0 as // we included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added // to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil (graph, path, pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ bool hamCycle(bool graph[V][V]) { int *path = new int[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if (hamCycleUtil(graph, path, 1) == false ) { cout << '
Solution does not exist'; return false; } printSolution(path); return true; } /* A utility function to print solution */ void printSolution(int path[]) { cout << 'Solution Exists:' ' Following is one Hamiltonian Cycle
'; for (int i = 0; i < V; i++) cout << path[i] << ' '; // Let us print the first vertex again // to show the complete cycle cout << path[0] << ' '; cout << endl; } // Driver Code int main() { /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ bool graph1[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}}; // Print the solution hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ bool graph2[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}}; // Print the solution hamCycle(graph2); return 0; } // This is code is contributed by rathbhupendra> C++ #include using namespace std; int main() { cout << 'GFG!'; return 0; }> C /* C program for solution of Hamiltonian Cycle problem using backtracking */ #include // Number of vertices in the graph #define V 5 void printSolution(int path[]); /* A utility function to check if the vertex v can be added at index 'pos' in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ int isSafe(int v, int graph[V][V], int path[], int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph [ path[pos-1] ][ v ] == 0) return 0; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return 0; return 1; } /* A recursive utility function to solve hamiltonian cycle problem */ int hamCycleUtil(int graph[V][V], int path[], int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the last included vertex to the // first vertex if ( graph[ path[pos-1] ][ path[0] ] == 1 ) return 1; else return 0; } // Try different vertices as a next candidate in Hamiltonian Cycle. // We don't try for 0 as we included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil (graph, path, pos+1) == 1) return 1; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return 0; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ int hamCycle(int graph[V][V]) { int path[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if ( hamCycleUtil(graph, path, 1) == 0 ) { printf('
Solution does not exist'); return 0; } printSolution(path); return 1; } /* A utility function to print solution */ void printSolution(int path[]) { printf ('Solution Exists:' ' Following is one Hamiltonian Cycle
'); for (int i = 0; i < V; i++) printf(' %d ', path[i]); // Let us print the first vertex again to show the complete cycle printf(' %d ', path[0]); printf('
'); } // driver program to test above function int main() { /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ int graph1[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}, }; // Print the solution hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ int graph2[V][V] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}, }; // Print the solution hamCycle(graph2); return 0; }> Java /* Java program for solution of Hamiltonian Cycle problem using backtracking */ class HamiltonianCycle { final int V = 5; int path[]; /* A utility function to check if the vertex v can be added at index 'pos'in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ boolean isSafe(int v, int graph[][], int path[], int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph[path[pos - 1]][v] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ boolean hamCycleUtil(int graph[][], int path[], int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the last included // vertex to the first vertex if (graph[path[pos - 1]][path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate in // Hamiltonian Cycle. We don't try for 0 as we // included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil(graph, path, pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ int hamCycle(int graph[][]) { path = new int[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if (hamCycleUtil(graph, path, 1) == false) { System.out.println('
Solution does not exist'); return 0; } printSolution(path); return 1; } /* A utility function to print solution */ void printSolution(int path[]) { System.out.println('Solution Exists: Following' + ' is one Hamiltonian Cycle'); for (int i = 0; i < V; i++) System.out.print(' ' + path[i] + ' '); // Let us print the first vertex again to show the // complete cycle System.out.println(' ' + path[0] + ' '); } // driver program to test above function public static void main(String args[]) { HamiltonianCycle hamiltonian = new HamiltonianCycle(); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ int graph1[][] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}, }; // Print the solution hamiltonian.hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ int graph2[][] = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}, }; // Print the solution hamiltonian.hamCycle(graph2); } } // This code is contributed by Abhishek Shankhadhar> Pytonorm # Python program for solution of # hamiltonian cycle problem class Graph(): def __init__(self, vertices): self.graph = [[0 for column in range(vertices)] for row in range(vertices)] self.V = vertices ''' Check if this vertex is an adjacent vertex of the previously added vertex and is not included in the path earlier ''' def isSafe(self, v, pos, path): # Check if current vertex and last vertex # in path are adjacent if self.graph[ path[pos-1] ][v] == 0: return False # Check if current vertex not already in path for vertex in path: if vertex == v: return False return True # A recursive utility function to solve # hamiltonian cycle problem def hamCycleUtil(self, path, pos): # base case: if all vertices are # included in the path if pos == self.V: # Last vertex must be adjacent to the # first vertex in path to make a cycle if self.graph[ path[pos-1] ][ path[0] ] == 1: return True else: return False # Try different vertices as a next candidate # in Hamiltonian Cycle. We don't try for 0 as # we included 0 as starting point in hamCycle() for v in range(1,self.V): if self.isSafe(v, pos, path) == True: path[pos] = v if self.hamCycleUtil(path, pos+1) == True: return True # Remove current vertex if it doesn't # lead to a solution path[pos] = -1 return False def hamCycle(self): path = [-1] * self.V ''' Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected ''' path[0] = 0 if self.hamCycleUtil(path,1) == False: print ('Solution does not exist
') return False self.printSolution(path) return True def printSolution(self, path): print ('Solution Exists: Following', 'is one Hamiltonian Cycle') for vertex in path: print (vertex ) # Driver Code ''' Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) ''' g1 = Graph(5) g1.graph = [ [0, 1, 0, 1, 0], [1, 0, 1, 1, 1], [0, 1, 0, 0, 1,],[1, 1, 0, 0, 1], [0, 1, 1, 1, 0], ] # Print the solution g1.hamCycle(); ''' Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) ''' g2 = Graph(5) g2.graph = [ [0, 1, 0, 1, 0], [1, 0, 1, 1, 1], [0, 1, 0, 0, 1,], [1, 1, 0, 0, 0], [0, 1, 1, 0, 0], ] # Print the solution g2.hamCycle(); # This code is contributed by Divyanshu Mehta> C# // C# program for solution of Hamiltonian // Cycle problem using backtracking using System; public class HamiltonianCycle { readonly int V = 5; int []path; /* A utility function to check if the vertex v can be added at index 'pos'in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ bool isSafe(int v, int [,]graph, int []path, int pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if (graph[path[pos - 1], v] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for (int i = 0; i < pos; i++) if (path[i] == v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ bool hamCycleUtil(int [,]graph, int []path, int pos) { /* base case: If all vertices are included in Hamiltonian Cycle */ if (pos == V) { // And if there is an edge from the last included // vertex to the first vertex if (graph[path[pos - 1],path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate in // Hamiltonian Cycle. We don't try for 0 as we // included 0 as starting point in hamCycle() for (int v = 1; v < V; v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe(v, graph, path, pos)) { path[pos] = v; /* recur to construct rest of the path */ if (hamCycleUtil(graph, path, pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ path[pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ int hamCycle(int [,]graph) { path = new int[V]; for (int i = 0; i < V; i++) path[i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ path[0] = 0; if (hamCycleUtil(graph, path, 1) == false) { Console.WriteLine('
Solution does not exist'); return 0; } printSolution(path); return 1; } /* A utility function to print solution */ void printSolution(int []path) { Console.WriteLine('Solution Exists: Following' + ' is one Hamiltonian Cycle'); for (int i = 0; i < V; i++) Console.Write(' ' + path[i] + ' '); // Let us print the first vertex again // to show the complete cycle Console.WriteLine(' ' + path[0] + ' '); } // Driver code public static void Main(String []args) { HamiltonianCycle hamiltonian = new HamiltonianCycle(); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ int [,]graph1= {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 1}, {0, 1, 1, 1, 0}, }; // Print the solution hamiltonian.hamCycle(graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ int [,]graph2 = {{0, 1, 0, 1, 0}, {1, 0, 1, 1, 1}, {0, 1, 0, 0, 1}, {1, 1, 0, 0, 0}, {0, 1, 1, 0, 0}, }; // Print the solution hamiltonian.hamCycle(graph2); } } // This code contributed by Rajput-Ji> Javascript >
PHP // PHP program for solution of // Hamiltonian Cycle problem // using backtracking $V = 5; /* A utility function to check if the vertex v can be added at index 'pos' in the Hamiltonian Cycle constructed so far (stored in 'path[]') */ function isSafe($v, $graph, &$path, $pos) { /* Check if this vertex is an adjacent vertex of the previously added vertex. */ if ($graph[$path[$pos - 1]][$v] == 0) return false; /* Check if the vertex has already been included. This step can be optimized by creating an array of size V */ for ($i = 0; $i < $pos; $i++) if ($path[$i] == $v) return false; return true; } /* A recursive utility function to solve hamiltonian cycle problem */ function hamCycleUtil($graph, &$path, $pos) { global $V; /* base case: If all vertices are included in Hamiltonian Cycle */ if ($pos == $V) { // And if there is an edge from the // last included vertex to the first vertex if ($graph[$path[$pos - 1]][$path[0]] == 1) return true; else return false; } // Try different vertices as a next candidate in // Hamiltonian Cycle. We don't try for 0 as we // included 0 as starting point hamCycle() for ($v = 1; $v < $V; $v++) { /* Check if this vertex can be added to Hamiltonian Cycle */ if (isSafe($v, $graph, $path, $pos)) { $path[$pos] = $v; /* recur to construct rest of the path */ if (hamCycleUtil($graph, $path, $pos + 1) == true) return true; /* If adding vertex v doesn't lead to a solution, then remove it */ $path[$pos] = -1; } } /* If no vertex can be added to Hamiltonian Cycle constructed so far, then return false */ return false; } /* This function solves the Hamiltonian Cycle problem using Backtracking. It mainly uses hamCycleUtil() to solve the problem. It returns false if there is no Hamiltonian Cycle possible, otherwise return true and prints the path. Please note that there may be more than one solutions, this function prints one of the feasible solutions. */ function hamCycle($graph) { global $V; $path = array_fill(0, $V, 0); for ($i = 0; $i < $V; $i++) $path[$i] = -1; /* Let us put vertex 0 as the first vertex in the path. If there is a Hamiltonian Cycle, then the path can be started from any point of the cycle as the graph is undirected */ $path[0] = 0; if (hamCycleUtil($graph, $path, 1) == false) { echo('
Solution does not exist'); return 0; } printSolution($path); return 1; } /* A utility function to print solution */ function printSolution($path) { global $V; echo('Solution Exists: Following is '. 'one Hamiltonian Cycle
'); for ($i = 0; $i < $V; $i++) echo(' '.$path[$i].' '); // Let us print the first vertex again to show the // complete cycle echo(' '.$path[0].'
'); } // Driver Code /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3)-------(4) */ $graph1 = array(array(0, 1, 0, 1, 0), array(1, 0, 1, 1, 1), array(0, 1, 0, 0, 1), array(1, 1, 0, 0, 1), array(0, 1, 1, 1, 0), ); // Print the solution hamCycle($graph1); /* Let us create the following graph (0)--(1)--(2) | / | | / | | / | (3) (4) */ $graph2 = array(array(0, 1, 0, 1, 0), array(1, 0, 1, 1, 1), array(0, 1, 0, 0, 1), array(1, 1, 0, 0, 0), array(0, 1, 1, 0, 0)); // Print the solution hamCycle($graph2); // This code is contributed by mits ?>> Produktion
Solution Exists: Following is one Hamiltonian Cycle 0 1 2 4 3 0 Solution does not exist>
Tidskomplexitet: O(N!), där N är antalet hörn.
Hjälputrymme: O(1), eftersom inget extra utrymme används.
Notera: Ovanstående kod skriver alltid ut en cykel som börjar från 0 . Startpunkten bör inte spela någon roll eftersom cykeln kan startas från vilken punkt som helst. Om du vill ändra startpunkten bör du göra två ändringar i ovanstående kod.
Ändra sökväg[0] = 0; till sökväg[0] = s ; var s är din nya startpunkt . Ändra också slinga för (int v = 1; v




