Skip to content

Consolidate duplicate implementations of max subarray #8849

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Jul 11, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 2 additions & 6 deletions DIRECTORY.md
Original file line number Diff line number Diff line change
Expand Up @@ -293,7 +293,7 @@
* [Inversions](divide_and_conquer/inversions.py)
* [Kth Order Statistic](divide_and_conquer/kth_order_statistic.py)
* [Max Difference Pair](divide_and_conquer/max_difference_pair.py)
* [Max Subarray Sum](divide_and_conquer/max_subarray_sum.py)
* [Max Subarray](divide_and_conquer/max_subarray.py)
* [Mergesort](divide_and_conquer/mergesort.py)
* [Peak](divide_and_conquer/peak.py)
* [Power](divide_and_conquer/power.py)
Expand Down Expand Up @@ -324,8 +324,7 @@
* [Matrix Chain Order](dynamic_programming/matrix_chain_order.py)
* [Max Non Adjacent Sum](dynamic_programming/max_non_adjacent_sum.py)
* [Max Product Subarray](dynamic_programming/max_product_subarray.py)
* [Max Sub Array](dynamic_programming/max_sub_array.py)
* [Max Sum Contiguous Subsequence](dynamic_programming/max_sum_contiguous_subsequence.py)
* [Max Subarray Sum](dynamic_programming/max_subarray_sum.py)
* [Min Distance Up Bottom](dynamic_programming/min_distance_up_bottom.py)
* [Minimum Coin Change](dynamic_programming/minimum_coin_change.py)
* [Minimum Cost Path](dynamic_programming/minimum_cost_path.py)
Expand Down Expand Up @@ -591,12 +590,10 @@
* [Is Square Free](maths/is_square_free.py)
* [Jaccard Similarity](maths/jaccard_similarity.py)
* [Juggler Sequence](maths/juggler_sequence.py)
* [Kadanes](maths/kadanes.py)
* [Karatsuba](maths/karatsuba.py)
* [Krishnamurthy Number](maths/krishnamurthy_number.py)
* [Kth Lexicographic Permutation](maths/kth_lexicographic_permutation.py)
* [Largest Of Very Large Numbers](maths/largest_of_very_large_numbers.py)
* [Largest Subarray Sum](maths/largest_subarray_sum.py)
* [Least Common Multiple](maths/least_common_multiple.py)
* [Line Length](maths/line_length.py)
* [Liouville Lambda](maths/liouville_lambda.py)
Expand Down Expand Up @@ -733,7 +730,6 @@
* [Linear Congruential Generator](other/linear_congruential_generator.py)
* [Lru Cache](other/lru_cache.py)
* [Magicdiamondpattern](other/magicdiamondpattern.py)
* [Maximum Subarray](other/maximum_subarray.py)
* [Maximum Subsequence](other/maximum_subsequence.py)
* [Nested Brackets](other/nested_brackets.py)
* [Number Container System](other/number_container_system.py)
Expand Down
112 changes: 112 additions & 0 deletions divide_and_conquer/max_subarray.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,112 @@
"""
The maximum subarray problem is the task of finding the continuous subarray that has the
maximum sum within a given array of numbers. For example, given the array
[-2, 1, -3, 4, -1, 2, 1, -5, 4], the contiguous subarray with the maximum sum is
[4, -1, 2, 1], which has a sum of 6.

This divide-and-conquer algorithm finds the maximum subarray in O(n log n) time.
"""
from __future__ import annotations

import time
from collections.abc import Sequence
from random import randint

from matplotlib import pyplot as plt


def max_subarray(
arr: Sequence[float], low: int, high: int
) -> tuple[int | None, int | None, float]:
"""
Solves the maximum subarray problem using divide and conquer.
:param arr: the given array of numbers
:param low: the start index
:param high: the end index
:return: the start index of the maximum subarray, the end index of the
maximum subarray, and the maximum subarray sum

>>> nums = [-2, 1, -3, 4, -1, 2, 1, -5, 4]
>>> max_subarray(nums, 0, len(nums) - 1)
(3, 6, 6)
>>> nums = [2, 8, 9]
>>> max_subarray(nums, 0, len(nums) - 1)
(0, 2, 19)
>>> nums = [0, 0]
>>> max_subarray(nums, 0, len(nums) - 1)
(0, 0, 0)
>>> nums = [-1.0, 0.0, 1.0]
>>> max_subarray(nums, 0, len(nums) - 1)
(2, 2, 1.0)
>>> nums = [-2, -3, -1, -4, -6]
>>> max_subarray(nums, 0, len(nums) - 1)
(2, 2, -1)
>>> max_subarray([], 0, 0)
(None, None, 0)
"""
if not arr:
return None, None, 0
if low == high:
return low, high, arr[low]

mid = (low + high) // 2
left_low, left_high, left_sum = max_subarray(arr, low, mid)
right_low, right_high, right_sum = max_subarray(arr, mid + 1, high)
cross_left, cross_right, cross_sum = max_cross_sum(arr, low, mid, high)
if left_sum >= right_sum and left_sum >= cross_sum:
return left_low, left_high, left_sum
elif right_sum >= left_sum and right_sum >= cross_sum:
return right_low, right_high, right_sum
return cross_left, cross_right, cross_sum


def max_cross_sum(
arr: Sequence[float], low: int, mid: int, high: int
) -> tuple[int, int, float]:
left_sum, max_left = float("-inf"), -1
right_sum, max_right = float("-inf"), -1

summ: int | float = 0
for i in range(mid, low - 1, -1):
summ += arr[i]
if summ > left_sum:
left_sum = summ
max_left = i

summ = 0
for i in range(mid + 1, high + 1):
summ += arr[i]
if summ > right_sum:
right_sum = summ
max_right = i

return max_left, max_right, (left_sum + right_sum)


def time_max_subarray(input_size: int) -> float:
arr = [randint(1, input_size) for _ in range(input_size)]
start = time.time()
max_subarray(arr, 0, input_size - 1)
end = time.time()
return end - start


def plot_runtimes() -> None:
input_sizes = [10, 100, 1000, 10000, 50000, 100000, 200000, 300000, 400000, 500000]
runtimes = [time_max_subarray(input_size) for input_size in input_sizes]
print("No of Inputs\t\tTime Taken")
for input_size, runtime in zip(input_sizes, runtimes):
print(input_size, "\t\t", runtime)
plt.plot(input_sizes, runtimes)
plt.xlabel("Number of Inputs")
plt.ylabel("Time taken in seconds")
plt.show()


if __name__ == "__main__":
"""
A random simulation of this algorithm.
"""
from doctest import testmod

testmod()
78 changes: 0 additions & 78 deletions divide_and_conquer/max_subarray_sum.py

This file was deleted.

93 changes: 0 additions & 93 deletions dynamic_programming/max_sub_array.py

This file was deleted.

60 changes: 60 additions & 0 deletions dynamic_programming/max_subarray_sum.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,60 @@
"""
The maximum subarray sum problem is the task of finding the maximum sum that can be
obtained from a contiguous subarray within a given array of numbers. For example, given
the array [-2, 1, -3, 4, -1, 2, 1, -5, 4], the contiguous subarray with the maximum sum
is [4, -1, 2, 1], so the maximum subarray sum is 6.

Kadane's algorithm is a simple dynamic programming algorithm that solves the maximum
subarray sum problem in O(n) time and O(1) space.

Reference: https://en.wikipedia.org/wiki/Maximum_subarray_problem
"""
from collections.abc import Sequence


def max_subarray_sum(
arr: Sequence[float], allow_empty_subarrays: bool = False
) -> float:
"""
Solves the maximum subarray sum problem using Kadane's algorithm.
:param arr: the given array of numbers
:param allow_empty_subarrays: if True, then the algorithm considers empty subarrays

>>> max_subarray_sum([2, 8, 9])
19
>>> max_subarray_sum([0, 0])
0
>>> max_subarray_sum([-1.0, 0.0, 1.0])
1.0
>>> max_subarray_sum([1, 2, 3, 4, -2])
10
>>> max_subarray_sum([-2, 1, -3, 4, -1, 2, 1, -5, 4])
6
>>> max_subarray_sum([2, 3, -9, 8, -2])
8
>>> max_subarray_sum([-2, -3, -1, -4, -6])
-1
>>> max_subarray_sum([-2, -3, -1, -4, -6], allow_empty_subarrays=True)
0
>>> max_subarray_sum([])
0
"""
if not arr:
return 0

max_sum = 0 if allow_empty_subarrays else float("-inf")
curr_sum = 0.0
for num in arr:
curr_sum = max(0 if allow_empty_subarrays else num, curr_sum + num)
max_sum = max(max_sum, curr_sum)

return max_sum


if __name__ == "__main__":
from doctest import testmod

testmod()

nums = [-2, 1, -3, 4, -1, 2, 1, -5, 4]
print(f"{max_subarray_sum(nums) = }")
Loading