Create largest_k_matrix.py

heaps/largest_k_matrix.py

@@ -0,0 +1,57 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+# author: bt3gl
+
+'''
+You are given an m x n binary matrix mat of 1's (representing soldiers) 
+and 0's (representing civilians). The soldiers are positioned in front 
+of the civilians. That is, all the 1's will appear to the left of
+all the 0's in each row.
+
+A row i is weaker than a row j if one of the following is true:
+
+The number of soldiers in row i is less than the number of soldiers in row j.
+Both rows have the same number of soldiers and i < j.
+Return the indices of the k weakest rows in the matrix ordered 
+from weakest to strongest.
+'''
+
+
+ def k_weakest_row(self, mat, k):
+
+        m = len(mat)
+        n = len(mat[0])
+
+        def binary_search(row):
+
+            low = 0
+            high = n
+
+            while low < high:
+
+                mid = (low + high) // 2
+                if row[mid] == 1:
+                    low = mid + 1
+                else:
+                    high = mid
+
+            return low
+
+
+        pq = []
+        for i, row in enumerate(mat):
+
+            strength = binary_search(row)
+            entry = (-strength, -i)
+            if len(pq) < k or entry > pq[0]:
+                heapq.heappush(pq, entry)
+            if len(pq) > k:
+                heapq.heappop(pq)
+            
+        indexes = []
+        while pq:
+            strength, i = heapq.heappop(pq)
+
+            indexes.append(-i)
+        
+        return indexes[::-1]