Design and implement a TwoSum class. It should support the following operations: add and find.
- add – Add the number to an internal data structure.
- find – Find if there exists any pair of numbers which sum is equal to the value.
Example 1:
add(1); add(3); add(5);
find(4) -> true
find(7) -> falseExample 2:
add(3); add(1); add(2);
find(3) -> true
find(6) -> false
Two-Sum is a very popular question to prepare for your coding interview. The essence to solve the two-sum question is to use a hash table/set.
Two Sum Interview Questions
- The Two Sum Algorithm using HashMap in C++/Java
- C++ Algorithms to Find Pair of Sum Given a Collection of Numbers
Using C++ std::unordered_map or map
Using a hash map to store the counter for each number, and do O(1) in inserting a new number to the list, and O(N) in finding a pair. The special case has to be handled for duplicate number pairs such as 5+5=10.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 | class TwoSum { public: /** Initialize your data structure here. */ TwoSum() { } /** Add the number to an internal data structure.. */ void add(int number) { data[number] ++; } /** Find if there exists any pair of numbers which sum is equal to the value. */ bool find(int value) { for (auto it = data.begin(); it != data.end(); it ++) { if (data.find(value - it->first) != data.end()) { if (it->first * 2 == value) { if (it->second > 1) { // special case return true; } } else { return true; } } } return false; } private: unordered_map<int, int> data; }; /** * Your TwoSum object will be instantiated and called as such: * TwoSum* obj = new TwoSum(); * obj->add(number); * bool param_2 = obj->find(value); */ |
class TwoSum { public: /** Initialize your data structure here. */ TwoSum() { } /** Add the number to an internal data structure.. */ void add(int number) { data[number] ++; } /** Find if there exists any pair of numbers which sum is equal to the value. */ bool find(int value) { for (auto it = data.begin(); it != data.end(); it ++) { if (data.find(value - it->first) != data.end()) { if (it->first * 2 == value) { if (it->second > 1) { // special case return true; } } else { return true; } } } return false; } private: unordered_map<int, int> data; }; /** * Your TwoSum object will be instantiated and called as such: * TwoSum* obj = new TwoSum(); * obj->add(number); * bool param_2 = obj->find(value); */
The above unordered_map can be replaced by std::map, however, it might be a bit slower as the map object will maintain its keys in ascending order i.e. O(logN) inserting, and internally, the MAP is implemented using a tree e.g. Red-Black tree while the unordered_map is a hash map.
C++ std::unordered_multiset or multiset
The multiset (the keys are sorted) or unordered_multiset in C++ allows you to insert duplicate numbers into the set. Therefore, by using the multiset, we can simplify the two-sum data structure by checking the counter of a number in the multiset.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 | class TwoSum { public: /** Initialize your data structure here. */ TwoSum() { } /** Add the number to an internal data structure.. */ void add(int number) { data.insert(number); } /** Find if there exists any pair of numbers which sum is equal to the value. */ bool find(int value) { for (const auto &n: data) { int c = value == n + n ? 1 : 0; if (data.count(value - n) > c) { return true; } } return false; } private: unordered_multiset<int> data; }; /** * Your TwoSum object will be instantiated and called as such: * TwoSum* obj = new TwoSum(); * obj->add(number); * bool param_2 = obj->find(value); */ |
class TwoSum { public: /** Initialize your data structure here. */ TwoSum() { } /** Add the number to an internal data structure.. */ void add(int number) { data.insert(number); } /** Find if there exists any pair of numbers which sum is equal to the value. */ bool find(int value) { for (const auto &n: data) { int c = value == n + n ? 1 : 0; if (data.count(value - n) > c) { return true; } } return false; } private: unordered_multiset<int> data; }; /** * Your TwoSum object will be instantiated and called as such: * TwoSum* obj = new TwoSum(); * obj->add(number); * bool param_2 = obj->find(value); */
The unordered_multiset is faster than multiset as the unordered version does not maintain the order of the keys like the multiset. The time complexity for inserting is O(1) and the find() takes O(N).
Vector and Two Pointer Algorithm
We can use a vector/array to store the numbers. For adding operation, we can use the std::upper_bound to find the position for inserting the element. This takes O(logN).
And as the elements are always sorted, we can use the two pointer algorithm that takes O(N) to find out if a sum pair exists in the vector.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | class TwoSum { public: /** Initialize your data structure here. */ TwoSum() { } /** Add the number to an internal data structure.. */ void add(int number) { auto it = upper_bound(begin(data), end(data), number); data.insert(it, number); } /** Find if there exists any pair of numbers which sum is equal to the value. */ bool find(int value) { int lo = 0, hi = data.size() - 1; while (lo < hi) { int mid = lo + (hi - lo) / 2; if (data[lo] + data[hi] == value) { return true; } if (data[lo] + data[hi] > value) { hi --; } else { lo ++; } } return false; } private: vector<int> data; }; /** * Your TwoSum object will be instantiated and called as such: * TwoSum* obj = new TwoSum(); * obj->add(number); * bool param_2 = obj->find(value); */ |
class TwoSum { public: /** Initialize your data structure here. */ TwoSum() { } /** Add the number to an internal data structure.. */ void add(int number) { auto it = upper_bound(begin(data), end(data), number); data.insert(it, number); } /** Find if there exists any pair of numbers which sum is equal to the value. */ bool find(int value) { int lo = 0, hi = data.size() - 1; while (lo < hi) { int mid = lo + (hi - lo) / 2; if (data[lo] + data[hi] == value) { return true; } if (data[lo] + data[hi] > value) { hi --; } else { lo ++; } } return false; } private: vector<int> data; }; /** * Your TwoSum object will be instantiated and called as such: * TwoSum* obj = new TwoSum(); * obj->add(number); * bool param_2 = obj->find(value); */
All the above Two-Sum data structures require O(N) space complexity to store the numbers.
Two Sum Variation Problems
- Teaching Kids Programming - Two Sum in Binary Search Tree via Inorder and Two Pointer Algorithm
- Teaching Kids Programming - Count Pairs Whose Sum is Less than Target (Two Pointer Algorithm)
- Teaching Kids Programming - Sum of Two Numbers Less Than Target using Two Pointer Algorithm
- Teaching Kids Programming - Two Pointer Algorithm to Solve Four Sum Problem
- Teaching Kids Programming - Recursive Algorithm to Find the Sum of Two Numbers in BSTs
- Two Pointer Algorithm to Count the Sum of Three Numbers Less than Target
- Recursive and Two Pointer Algorithms to Determine Four Sum
- Algorithms to Check Sum of Two Numbers in Binary Search Trees
- Teaching Kids Programming - 3 Different Approaches to Solve Two-Sum Problem
- How to Design a Two-Sum Data Structure?
- How to Find the Closest Sum of Three in an Array using Two Pointer Algorithm? (3Sum Closest)
- Teaching Kids Programming - Three Sum Algorithm
- Teaching Kids Programming – Two Sum Algorithm when Input Array is Sorted
- Teaching Kids Programming – Two Sum Algorithm
- Two Pointer Algorithm to Find Maximum Two Sum Less Than K
- The Two Sum Algorithm using HashMap in C++/Java
–EOF (The Ultimate Computing & Technology Blog) —
loading...
Last Post: The MySQL If and Case Statement Exercise: How to Swap Elements in MySQL?
Next Post: How to Design a Snake Game?