2021-08-24

class Solution:
    def maximalSquare(self, matrix) :
        row = len(matrix)
        col = len(matrix[0])
        res = [[0 for _ in range(col)] for _ in range(row)]
        ans = 0
        for i in range(row):
            for j in range(col):
                if matrix[i][j]=='1':
                    if 0<=i-1<row and 0<=j-1<col:
                        if res[i-1][j]==res[i][j-1]:
                            if matrix[i-1-res[i-1][j]+1][j-res[i-1][j]] == '1':
                                res[i][j] = res[i-1][j]+1
                            else:
                                res[i][j] = res[i - 1][j]
                        else:
                            res[i][j] = min(res[i-1][j],res[i][j-1])+1
                    else:
                        res[i][j] +=1

        print(res,max(res),max(max(res)))
        # return np.max(res)**2
        # return max(map(max, res))**2
        return max(max(row) for row in res)**2

• 思路
  动态规划，利用res[i-1][j]和res[i][j-1]确定res[i][j]
• 趣点
  二维list求最大值

1. 利用map，max(map(max, res))
2. 按列循环求解，max(max(row) for rol in res), 相比1速度更快，内存更小
3. 利用numpy， np.max(res)，在leetcode上通不过就不测了