A typical O(n^2) solution uses dynamic programming. Let's use lens[j] to denote the length of the LIS ending with nums[j]. The state equations are
lens[0] = 1
lens[j] = max_{i = 0, 1, 2, ..., j - 1}(lens[j], lens[i] + 1)
Then the length of the LIS of nums is just the maximum value in lens. The code is as follows.
1 class Solution { 2 public: 3 int lengthOfLIS(vector<int>& nums) { 4 if (nums.empty()) return 0; 5 int n = nums.size(), ml = 1; 6 vector<int> lens(n, 1); 7 for (int j = 1; j < n; j++) { 8 for (int i = 0; i < j; i++) 9 if (nums[j] > nums[i]) 10 lens[j] = max(lens[j], lens[i] + 1); 11 ml = max(ml, lens[j]); 12 } 13 return ml; 14 } 15 };
The O(nlogn) solution is much more complicated. Try to read through the explanation of it in GeeksforGeeks first. The code is as follows.
class Solution { public: int lengthOfLIS(vector<int>& nums) { if (nums.empty()) return 0; vector<int> ends{nums[0]}; for (int num : nums) { if (num < ends[0]) ends[0] = num; else if (num > ends.back()) ends.push_back(num); else { int l = 0, r = ends.size() - 1; while (l < r) { int m = l + (r - l) / 2; if (ends[m] < num) l = m + 1; else r = m; } ends[r] = num; } } return ends.size(); } };
If you look at the else part carefully, you may notice that it can be done using lower_bound. So we will have a much shorter code, like this one.
1 class Solution { 2 public: 3 int lengthOfLIS(vector<int>& nums) { 4 vector<int> ends; 5 for (int num : nums) { 6 auto it = lower_bound(ends.begin(), ends.end(), num); 7 if (it == ends.end()) ends.push_back(num); 8 else *it = num; 9 } 10 return ends.size(); 11 } 12 };