High Frequency Questions

This section will record all the popular questions and their follow ups during the interviews.

LeetCode 347. Top K Frequent Elements

It seems that the initial problem is not hard, you can have 2 solutions:

• Using minHeap which is O(nlogk)
• Using BucketSort which is O(n)

1. Solution with Dict + Sort

The code below is using bucket sort, and i’m using the lambda function to simplify the code::

class Solution(object):

def topKFrequent(self, nums, k):

“”” :type nums: List[int] :type k: int :rtype: List[int] “”” # if we do a mergesort, the time completxisty will be O(nlogn) res = dict()

for num in nums:

if num in res:

res[num] += 1

else:

res[num] = 1

# lambda is the most used functions during my work res_sorted = sorted(res.items(), key=lambda x: x[1]) return [item[0] for item in res_sorted[-1:-1 - k:-1]]

# I can’t believe i’m so fast that beats 93% of solutions!!!

1. Solution with Heap
2. Follow up–how do you update the K top results in real time?