I have been working on the LeetCode problem 5. Longest Palindromic Substring:
Given a string s, return the longest palindromic substring in s.
But I kept getting time limit exceeded on large test cases.
I used dynamic programming as follows:
dp[(i, j)] = True implies that s[i] to s[j] is a palindrome. So if s[i] == str[j] and dp[(i+1, j-1]) is set to True, that means S[i] to S[j] is also a palindrome.
How can I improve the performance of this implementation?
class Solution:
def longestPalindrome(self, s: str) -> str:
dp = {}
res = ""
for i in range(len(s)):
# single character is always a palindrome
dp[(i, i)] = True
res = s[i]
#fill in the table diagonally
for x in range(len(s) - 1):
i = 0
j = x + 1
while j <= len(s)-1:
if s[i] == s[j] and (j - i == 1 or dp[(i+1, j-1)] == True):
dp[(i, j)] = True
if(j-i+1) > len(res):
res = s[i:j+1]
else:
dp[(i, j)] = False
i += 1
j += 1
return res
I think the judging system for this problem is kind of too tight, it took some time to make it pass, improved version:
class Solution:
def longestPalindrome(self, s: str) -> str:
dp = {}
res = ""
for i in range(len(s)):
dp[(i, i)] = True
res = s[i]
for x in range(len(s)): # iterate till the end of the string
for i in range(x): # iterate up to the current state (less work) and for loop looks better here
if s[i] == s[x] and (dp.get((i + 1, x - 1), False) or x - i == 1):
dp[(i, x)] = True
if x - i + 1 > len(res):
res = s[i:x + 1]
return res
Here is another idea to improve the performance:
The nested loop will check over many cases where the DP value is already False for smaller ranges. We can avoid looking at large spans, by looking for palindromes from inside-out and stop extending the span as soon as it no longer is a palindrome. This process should be repeated at every offset in the source string, but this could still save some processing.
The inputs for which then most time is wasted, are those where there are lots of the same letters after each other, like "aaaaaaabcaaaaaaa". These lead to many iterations: each "a" or "aa" could be the center of a palindrome, but "growing" each of them is a waste of time. We should just consider all consecutive "a" together from the start and expand from there onwards.
You can specifically deal with these cases by first grouping consecutive letters which are the same. So the above example would be turned into 4 groups: a(7)b(1)c(1)a(7)
Then let each group in turn be taken as the center of a palindrome. For each group, "fan out" to potentially include one or more neighboring groups at both sides in "tandem". Continue fanning out until either the outside groups are not about the same letter, or they have a different group size. From that result you can derive what the largest palindrome is around that center. In particular, when the case is that the letters of the outer groups are the same, but not their sizes, you still include that letter at the outside of the palindrome, but with a repetition that corresponds to the least of these two mismatching group sizes.
Here is an implementation. I used named tuples to make it more readable:
from itertools import groupby
from collections import namedtuple
Group = namedtuple("Group", "letter,size,end")
class Solution:
def longestPalindrome(self, s: str) -> str:
longest = ""
x = 0
groups = [Group(group[0], len(group), x := x + len(group)) for group in
("".join(group[1]) for group in groupby(s))]
for i in range(len(groups)):
for j in range(0, min(i+1, len(groups) - i)):
if groups[i - j].letter != groups[i + j].letter:
break
left = groups[i - j]
right = groups[i + j]
if left.size != right.size:
break
size = right.end - (left.end - left.size) - abs(left.size - right.size)
if size > len(longest):
x = left.end - left.size + max(0, left.size - right.size)
longest = s[x:x+size]
return longest
Alternatively, you can try this approach, it seems to be faster than 96% Python submission.
def longestPalindrome(self, s: str) -> str:
N = len(s)
if N == 0:
return 0
max_len, start = 1, 0
for i in range(N):
df = i - max_len
if df >= 1 and s[df-1: i+1] == s[df-1: i+1][::-1]:
start = df - 1
max_len += 2
continue
if df >= 0 and s[df: i+1] == s[df: i+1][::-1]:
start= df
max_len += 1
return s[start: start + max_len]
If you want to improve the performance, you should create a variable for len(s) at the beginning of the function and use it. That way instead of calling len(s) 3 times, you would do it just once.
Also, I see no reason to create a class for this function. A simple function will outrun a class method, albeit very slightly.
I tried this to count the number of occurences of the target in a string in a recursive way, but I don't think where len(target) > 1 is a recursive way. I can't really think of other way to do this in a recursive way. I am doing this without using any string methods except indexing and slicing. Please give some help.
target can be a single character or a substring.
For example, if s = 'aaaaabb' and target = 'aa', i want the output to be 4.
def count(s, target):
if s == '':
return 0
if len(target) > len(s):
return 0
if len(target) <= 1:
if s[0] == target:
return 1 + count(s[1:], target)
else:
return 0 + count(s[1:], target)
if len(target) > 1:
count = 0
for x in range(len(s) - len(target) + 1):
if s[x:x+len(target)] == target:
count += 1
return count
Try:
def count_substring(s, target):
if not s:
return 0
return (s[:len(target)] == target) + count_substring(s[1:], target)
print(count_substring('aaaaabb', 'aa')) # 4
The idea is that at each recursion, the function only cares about the left-most substring, to compare with target. You can think of it as a slight modification of your len(target) <= 1 case.
I've been doing a good amount of studying and practicing algorithms, and I ran into one that asked the question (summarizing here) "Given two strings, return True if the strings are one edit away (either by removing, inserting, or replacing a character). Return False if not."
I went about this problem by comparing two strings and counting the amount of letters that are in string1, but not in string2. If there is more than one letter missing, then it will return False.
Here is my Python code:
def oneAway(string1, string2):
string1 = string1.lower()
string2 = string2.lower()
# counts the number of edits required
counter = 0
for i in string1:
if i not in string2:
counter += 1
if counter > 1:
return False
else:
return True
I'd like to hear other people's approaches to this problem, and please point out if I have oversimplified this concept.
Why call .lower() on both strings? Based on the question, oneAway('abc', 'ABC') should be False.
Building off of the other comments, how about this:
def oneAway(s1, s2):
i, j = 0, 0
mistake_occurred = False
while i < len(s1) and j < len(s2):
if s1[i] != s2[j]:
if mistake_occurred:
return False
mistake_occurred = True
i += 1
j += 1
if mistake_occurred and (i != len(s1) or j != len(s2)):
return False
if i < len(s1) - 1 or j < len(s2) - 1:
return False
return True
You have to check each editing action specifically:
def isOneEdit(A,B):
# replacing one char
if len(A) == len(B) and sum(a!=b for a,b in zip(A,B)) == 1:
return True
# inserting one char in A to get B
if len(A) == len(B)-1 and any(A==B[:i]+B[i+1:] for i in range(len(B))):
return True
# removing one char (like inserting one in B to get A)
if len(A) == len(B)+1:
return isOneEdit(B,A)
return False
print(isOneEdit("abc","abd")) # True - replace c with d
print(isOneEdit("abd","abcd")) # True - insert c
print(isOneEdit("abcd","abd")) # True - delete c
print(isOneEdit("abcde","abd")) # False
Alternatively, you can compare the size of the common prefix and suffix of the two strings to the length of the longest one:
def isOneEdit(A,B):
if abs(len(A)-len(B))>1: return False
commonPrefix = next((i for i,(a,b) in enumerate(zip(A,B)) if a!=b),len(A))
commonSuffix = next((i for i,(a,b) in enumerate(zip(reversed(A),reversed(B))) if a!=b),len(B))
editSize = max(len(A),len(B)) - (commonPrefix+commonSuffix)
return editSize <= 1
I am trying to implement the binary search in python and have written it as follows. However, I can't make it stop whenever needle_element is larger than the largest element in the array.
Can you help? Thanks.
def binary_search(array, needle_element):
mid = (len(array)) / 2
if not len(array):
raise "Error"
if needle_element == array[mid]:
return mid
elif needle_element > array[mid]:
return mid + binary_search(array[mid:],needle_element)
elif needle_element < array[mid]:
return binary_search(array[:mid],needle_element)
else:
raise "Error"
It would be much better to work with a lower and upper indexes as Lasse V. Karlsen was suggesting in a comment to the question.
This would be the code:
def binary_search(array, target):
lower = 0
upper = len(array)
while lower < upper: # use < instead of <=
x = lower + (upper - lower) // 2
val = array[x]
if target == val:
return x
elif target > val:
if lower == x: # these two are the actual lines
break # you're looking for
lower = x
elif target < val:
upper = x
lower < upper will stop once you have reached the smaller number (from the left side)
if lower == x: break will stop once you've reached the higher number (from the right side)
Example:
>>> binary_search([1,5,8,10], 5) # return 1
1
>>> binary_search([1,5,8,10], 0) # return None
>>> binary_search([1,5,8,10], 15) # return None
Why not use the bisect module? It should do the job you need---less code for you to maintain and test.
array[mid:] creates a new sub-copy everytime you call it = slow. Also you use recursion, which in Python is slow, too.
Try this:
def binarysearch(sequence, value):
lo, hi = 0, len(sequence) - 1
while lo <= hi:
mid = (lo + hi) // 2
if sequence[mid] < value:
lo = mid + 1
elif value < sequence[mid]:
hi = mid - 1
else:
return mid
return None
In the case that needle_element > array[mid], you currently pass array[mid:] to the recursive call. But you know that array[mid] is too small, so you can pass array[mid+1:] instead (and adjust the returned index accordingly).
If the needle is larger than all the elements in the array, doing it this way will eventually give you an empty array, and an error will be raised as expected.
Note: Creating a sub-array each time will result in bad performance for large arrays. It's better to pass in the bounds of the array instead.
You can improve your algorithm as the others suggested, but let's first look at why it doesn't work:
You're getting stuck in a loop because if needle_element > array[mid], you're including element mid in the bisected array you search next. So if needle is not in the array, you'll eventually be searching an array of length one forever. Pass array[mid+1:] instead (it's legal even if mid+1 is not a valid index), and you'll eventually call your function with an array of length zero. So len(array) == 0 means "not found", not an error. Handle it appropriately.
This is a tail recursive solution, I think this is cleaner than copying partial arrays and then keeping track of the indexes for returning:
def binarySearch(elem, arr):
# return the index at which elem lies, or return false
# if elem is not found
# pre: array must be sorted
return binarySearchHelper(elem, arr, 0, len(arr) - 1)
def binarySearchHelper(elem, arr, start, end):
if start > end:
return False
mid = (start + end)//2
if arr[mid] == elem:
return mid
elif arr[mid] > elem:
# recurse to the left of mid
return binarySearchHelper(elem, arr, start, mid - 1)
else:
# recurse to the right of mid
return binarySearchHelper(elem, arr, mid + 1, end)
def binary_search(array, target):
low = 0
mid = len(array) / 2
upper = len(array)
if len(array) == 1:
if array[0] == target:
print target
return array[0]
else:
return False
if target == array[mid]:
print array[mid]
return mid
else:
if mid > low:
arrayl = array[0:mid]
binary_search(arrayl, target)
if upper > mid:
arrayu = array[mid:len(array)]
binary_search(arrayu, target)
if __name__ == "__main__":
a = [3,2,9,8,4,1,9,6,5,9,7]
binary_search(a,9)
Using Recursion:
def binarySearch(arr,item):
c = len(arr)//2
if item > arr[c]:
ans = binarySearch(arr[c+1:],item)
if ans:
return binarySearch(arr[c+1],item)+c+1
elif item < arr[c]:
return binarySearch(arr[:c],item)
else:
return c
binarySearch([1,5,8,10,20,50,60],10)
All the answers above are true , but I think it would help to share my code
def binary_search(number):
numbers_list = range(20, 100)
i = 0
j = len(numbers_list)
while i < j:
middle = int((i + j) / 2)
if number > numbers_list[middle]:
i = middle + 1
else:
j = middle
return 'the index is '+str(i)
If you're doing a binary search, I'm guessing the array is sorted. If that is true you should be able to compare the last element in the array to the needle_element. As octopus says, this can be done before the search begins.
You can just check to see that needle_element is in the bounds of the array before starting at all. This will make it more efficient also, since you won't have to do several steps to get to the end.
if needle_element < array[0] or needle_element > array[-1]:
# do something, raise error perhaps?
It returns the index of key in array by using recursive.
round() is a function convert float to integer and make code fast and goes to expected case[O(logn)].
A=[1,2,3,4,5,6,7,8,9,10]
low = 0
hi = len(A)
v=3
def BS(A,low,hi,v):
mid = round((hi+low)/2.0)
if v == mid:
print ("You have found dude!" + " " + "Index of v is ", A.index(v))
elif v < mid:
print ("Item is smaller than mid")
hi = mid-1
BS(A,low,hi,v)
else :
print ("Item is greater than mid")
low = mid + 1
BS(A,low,hi,v)
BS(A,low,hi,v)
Without the lower/upper indexes this should also do:
def exists_element(element, array):
if not array:
yield False
mid = len(array) // 2
if element == array[mid]:
yield True
elif element < array[mid]:
yield from exists_element(element, array[:mid])
else:
yield from exists_element(element, array[mid + 1:])
Returning a boolean if the value is in the list.
Capture the first and last index of the list, loop and divide the list capturing the mid value.
In each loop will do the same, then compare if value input is equal to mid value.
def binarySearch(array, value):
array = sorted(array)
first = 0
last = len(array) - 1
while first <= last:
midIndex = (first + last) // 2
midValue = array[midIndex]
if value == midValue:
return True
if value < midValue:
last = midIndex - 1
if value > midValue:
first = midIndex + 1
return False
I'm trying to make a slightly more advanced palindrome as according to the docstring. However I can't get it to work. Am I going it about the right way? This is what I have so far:
def pal_length(s: str, n: int) -> bool:
'''Return True iff s has a palindrome of length exactly n.
>>> pal_length('abclevel', 5)
True
>>> pal_length('level', 2)
False
'''
if not s:
return True
else:
index = 0
while index < len(s):
if s[index] == s[index+n]:
return pal_length(s[index+1:index+n-1],n-1)
index += 1
return False
I'm trying to not use any import modules etc. Just straight recursion.
Any help is appreciated. Thanks.
I think your indexing is a bit off. Shouldn't it be
index = 0
while index < len(s) - n + 1:
if s[index] == s[index+n-1]:
return pal_length(s[index+1:index+n-1], n-2)
index += 1
return False