矩阵对角线元素的和 - 简单

*************

c++

topic: 1572. 矩阵对角线元素的和 - 力扣(LeetCode)

*************

Look at the problems immediately.

|----------------------------------------------------------------------------|
| |

vector<vector<int>>& mat means mat is a two-dimension vector. Let's review the basic usage of the creating vector in c++.

make an integer.

cpp 复制代码
int w = 13;
int t = 38;

make a one-dimension vector.

cpp 复制代码
// 直接给定数组,数组的名字是自定义的
vector<int> w = {1, 3, 3, 8};

// 构造一个数组,包含13个元素,每个元素是 38
vector<int> t(13, 38);

make a two-dimension vector. And when talks about two-dimension vector, it is made of many one-dimension vctors.

cpp 复制代码
// 一维数组
vecotr<int> w(13, 38);

输出:
38 38 38 38 38 38 38 38 38 38 38 38 38


// 二维数组就是规定了有几个一维数组、
vector<vector<int>> t(13, vector<int>(13, 38));

输出:
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38
38 38 38 38 38 38 38 38 38 38 38 38 38

I like the basic usages of everything so much. Making full usage of the things keeps claen. Many people want to learn too much skills, which I think donnot have to. Keep things simple.

I think when looking at the mat, getting the size is always first.

cpp 复制代码
class Solution {
public:
    int diagonalSum(vector<vector<int>>& mat) {
        
        int n = mat.size(); // get the size of mat
        int sum = 0;
    }
};

mat[a][b] means the element lies in line a column b.

cpp 复制代码
class Solution {
public:
    int diagonalSum(vector<vector<int>>& mat) {
        
        int n = mat.size(); // get the size of mat
        int sum = 0;

        for (int i = 0; i < n; i++)
        {
            sum = sum + mat[i][i];
            sum = sum + mat[i][n - 1 - i];
        }

        return sum;
    }
};

|----------------------------------------------------------------------------|
| |

This problen is easy but sumething wrong. Soon I find the key point. 5 is really a special one. It lies in both main diagonal and counter diagonal.

|----------------------------------------------------------------------------|
| |

just minus it.

cpp 复制代码
class Solution {
public:
    int diagonalSum(vector<vector<int>>& mat) {
        
        int n = mat.size(); // get the size of mat
        int sum = 0;

        for (int i = 0; i < n; i++)
        {
            sum = sum + mat[i][i];
            sum = sum + mat[i][n - 1 - i];
        }

        // 如果奇数个元素,那么得减掉正中心的元素,因为他被计算了两遍
        if (n % 2 == 1)
        {
            sum = sum - mat[(n - 1) / 2][(n - 1) / 2];
        }

        return sum;
    }
};

|----------------------------------------------------------------------------|
| |

相关推荐
_OP_CHEN9 小时前
【算法基础篇】(五十七)线性代数之矩阵乘法从入门到实战:手撕模板 + 真题详解
线性代数·算法·矩阵·蓝桥杯·c/c++·矩阵乘法·acm/icpc
芷栀夏10 小时前
CANN ops-math:从矩阵运算到数值计算的全维度硬件适配与效率提升实践
人工智能·神经网络·线性代数·矩阵·cann
种时光的人19 小时前
CANN仓库核心解读:catlass夯实AIGC大模型矩阵计算的算力基石
线性代数·矩阵·aigc
Zfox_1 天前
CANN Catlass 算子模板库深度解析:高性能矩阵乘(GEMM)原理、融合优化与模板化开发实践
线性代数·矩阵
lbb 小魔仙1 天前
面向 NPU 的高性能矩阵乘法:CANN ops-nn 算子库架构与优化技术
线性代数·矩阵·架构
空白诗1 天前
CANN ops-nn 算子解读:大语言模型推理中的 MatMul 矩阵乘实现
人工智能·语言模型·矩阵
劈星斩月1 天前
线性代数-3Blue1Brown《线性代数的本质》特征向量与特征值(12)
线性代数·特征值·特征向量·特征方程
池央1 天前
ops-nn 算子库中的数据布局与混合精度策略:卷积、矩阵乘法与 RNN 的优化实践
rnn·线性代数·矩阵
深鱼~1 天前
大模型底层算力支撑:ops-math在矩阵乘法上的优化
人工智能·线性代数·矩阵·cann
Zfox_1 天前
CANN PyPTO 编程范式深度解析:并行张量与 Tile 分块操作的架构原理、内存控制与流水线调度机制
线性代数·矩阵·架构