Adott sok köteg érme, amelyek egymás mellett vannak elrendezve. Ezeket az érméket minimális számú lépésben kell összegyűjtenünk, ahol egy lépésben egy vízszintes vagy függőleges érmesort gyűjthetünk össze, és az összegyűjtött érméknek folyamatosnak kell lenniük.
Példák:
Input : height[] = [2 1 2 5 1] Each value of this array corresponds to the height of stack that is we are given five stack of coins where in first stack 2 coins are there then in second stack 1 coin is there and so on. Output : 4 We can collect all above coins in 4 steps which are shown in below diagram. Each step is shown by different color. First we have collected last horizontal line of coins after which stacks remains as [1 0 1 4 0] after that another horizontal line of coins is collected from stack 3 and 4 then a vertical line from stack 4 and at the end a horizontal line from stack 1. Total steps are 4.
nevezzen át egy mappát Linuxon
Ezt a problémát az oszd meg és uralkodj módszerrel tudjuk megoldani. Láthatjuk, hogy mindig előnyös a vízszintes vonalak alulról történő eltávolítása. Tegyük fel, hogy az l indextől az r indexig tartó veremeken dolgozunk egy rekurziós lépésben minden alkalommal, amikor kiválasztjuk a minimális magasságot, távolítsuk el azt a sok vízszintes vonalat, amely után a verem két részre lesz bontva l-től minimumig és minimum +1-ig és r-ig, és rekurzív módon hívjuk meg ezeket az altömböket. Egy másik dolog, hogy függőleges vonalak használatával is gyűjthetünk érméket, így a minimumot választjuk a rekurzív hívások eredménye és az (r - l) között, mivel az (r - l) függőleges vonalak használatával mindig összegyűjthetjük az összes érmét.
Ahogy minden egyes altömb meghívásakor megtaláljuk a megoldás teljes időbonyolultságának minimumát, O(N2)
C++
// C++ program to find minimum number of // steps to collect stack of coins #include using namespace std; // recursive method to collect coins from // height array l to r with height h already // collected int minStepsRecur(int height[] int l int r int h) { // if l is more than r no steps needed if (l >= r) return 0; // loop over heights to get minimum height // index int m = l; for (int i = l; i < r; i++) if (height[i] < height[m]) m = i; /* choose minimum from 1) collecting coins using all vertical lines (total r - l) 2) collecting coins using lower horizontal lines and recursively on left and right segments */ return min(r - l minStepsRecur(height l m height[m]) + minStepsRecur(height m + 1 r height[m]) + height[m] - h); } // method returns minimum number of step to // collect coin from stack with height in // height[] array int minSteps(int height[] int N) { return minStepsRecur(height 0 N 0); } // Driver code to test above methods int main() { int height[] = { 2 1 2 5 1 }; int N = sizeof(height) / sizeof(int); cout << minSteps(height N) << endl; return 0; }
Java // Java Code to Collect all coins in // minimum number of steps import java.util.*; class GFG { // recursive method to collect coins from // height array l to r with height h already // collected public static int minStepsRecur(int height[] int l int r int h) { // if l is more than r no steps needed if (l >= r) return 0; // loop over heights to get minimum height // index int m = l; for (int i = l; i < r; i++) if (height[i] < height[m]) m = i; /* choose minimum from 1) collecting coins using all vertical lines (total r - l) 2) collecting coins using lower horizontal lines and recursively on left and right segments */ return Math.min(r - l minStepsRecur(height l m height[m]) + minStepsRecur(height m + 1 r height[m]) + height[m] - h); } // method returns minimum number of step to // collect coin from stack with height in // height[] array public static int minSteps(int height[] int N) { return minStepsRecur(height 0 N 0); } /* Driver program to test above function */ public static void main(String[] args) { int height[] = { 2 1 2 5 1 }; int N = height.length; System.out.println(minSteps(height N)); } } // This code is contributed by Arnav Kr. Mandal.
Python 3 # Python 3 program to find # minimum number of steps # to collect stack of coins # recursive method to collect # coins from height array l to # r with height h already # collected def minStepsRecur(height l r h): # if l is more than r # no steps needed if l >= r: return 0; # loop over heights to # get minimum height index m = l for i in range(l r): if height[i] < height[m]: m = i # choose minimum from # 1) collecting coins using # all vertical lines (total r - l) # 2) collecting coins using # lower horizontal lines and # recursively on left and # right segments return min(r - l minStepsRecur(height l m height[m]) + minStepsRecur(height m + 1 r height[m]) + height[m] - h) # method returns minimum number # of step to collect coin from # stack with height in height[] array def minSteps(height N): return minStepsRecur(height 0 N 0) # Driver code height = [ 2 1 2 5 1 ] N = len(height) print(minSteps(height N)) # This code is contributed # by ChitraNayal
C# // C# Code to Collect all coins in // minimum number of steps using System; class GFG { // recursive method to collect coins from // height array l to r with height h already // collected public static int minStepsRecur(int[] height int l int r int h) { // if l is more than r no steps needed if (l >= r) return 0; // loop over heights to // get minimum height index int m = l; for (int i = l; i < r; i++) if (height[i] < height[m]) m = i; /* choose minimum from 1) collecting coins using all vertical lines (total r - l) 2) collecting coins using lower horizontal lines and recursively on left and right segments */ return Math.Min(r - l minStepsRecur(height l m height[m]) + minStepsRecur(height m + 1 r height[m]) + height[m] - h); } // method returns minimum number of step to // collect coin from stack with height in // height[] array public static int minSteps(int[] height int N) { return minStepsRecur(height 0 N 0); } /* Driver program to test above function */ public static void Main() { int[] height = { 2 1 2 5 1 }; int N = height.Length; Console.Write(minSteps(height N)); } } // This code is contributed by nitin mittal
PHP // PHP program to find minimum number of // steps to collect stack of coins // recursive method to collect // coins from height array l to // r with height h already // collected function minStepsRecur($height $l $r $h) { // if l is more than r // no steps needed if ($l >= $r) return 0; // loop over heights to // get minimum height // index $m = $l; for ($i = $l; $i < $r; $i++) if ($height[$i] < $height[$m]) $m = $i; /* choose minimum from 1) collecting coins using all vertical lines (total r - l) 2) collecting coins using lower horizontal lines and recursively on left and right segments */ return min($r - $l minStepsRecur($height $l $m $height[$m]) + minStepsRecur($height $m + 1 $r $height[$m]) + $height[$m] - $h); } // method returns minimum number of step to // collect coin from stack with height in // height[] array function minSteps($height $N) { return minStepsRecur($height 0 $N 0); } // Driver Code $height = array(2 1 2 5 1); $N = sizeof($height); echo minSteps($height $N) ; // This code is contributed by nitin mittal. ?> JavaScript <script> // Javascript Code to Collect all coins in // minimum number of steps // recursive method to collect coins from // height array l to r with height h already // collected function minStepsRecur(heightlrh) { // if l is more than r no steps needed if (l >= r) return 0; // loop over heights to get minimum height // index let m = l; for (let i = l; i < r; i++) if (height[i] < height[m]) m = i; /* choose minimum from 1) collecting coins using all vertical lines (total r - l) 2) collecting coins using lower horizontal lines and recursively on left and right segments */ return Math.min(r - l minStepsRecur(height l m height[m]) + minStepsRecur(height m + 1 r height[m]) + height[m] - h); } // method returns minimum number of step to // collect coin from stack with height in // height[] array function minSteps(heightN) { return minStepsRecur(height 0 N 0); } /* Driver program to test above function */ let height=[2 1 2 5 1 ]; let N = height.length; document.write(minSteps(height N)); // This code is contributed by avanitrachhadiya2155 </script>
Kimenet:
4
Időbeli összetettség: Ennek az algoritmusnak az időbonyolultsága O(N^2), ahol N a magassági tömb elemeinek száma.
A tér összetettsége: Ennek az algoritmusnak a térbeli összetettsége O(N) a magassági tömbön végrehajtott rekurzív hívások miatt.