如何生成列表的所有排列?例如:
permutations([])
[]
permutations([1])
[1]
permutations([1, 2])
[1, 2]
[2, 1]
permutations([1, 2, 3])
[1, 2, 3]
[1, 3, 2]
[2, 1, 3]
[2, 3, 1]
[3, 1, 2]
[3, 2, 1]
如何生成列表的所有排列?例如:
permutations([])
[]
permutations([1])
[1]
permutations([1, 2])
[1, 2]
[2, 1]
permutations([1, 2, 3])
[1, 2, 3]
[1, 3, 2]
[2, 1, 3]
[2, 3, 1]
[3, 1, 2]
[3, 2, 1]
当前回答
对于Python 2.6及以上版本:
import itertools
itertools.permutations([1, 2, 3])
这将作为生成器返回。使用列表(排列(xs))作为列表返回。
其他回答
def permutations(head, tail=''):
if len(head) == 0:
print(tail)
else:
for i in range(len(head)):
permutations(head[:i] + head[i+1:], tail + head[i])
称为:
permutations('abc')
以下代码是给定列表的就地排列,作为生成器实现。由于它只返回对列表的引用,因此不应在生成器外部修改列表。该解决方案是非递归的,因此使用了低内存。还可以很好地处理输入列表中元素的多个副本。
def permute_in_place(a):
a.sort()
yield list(a)
if len(a) <= 1:
return
first = 0
last = len(a)
while 1:
i = last - 1
while 1:
i = i - 1
if a[i] < a[i+1]:
j = last - 1
while not (a[i] < a[j]):
j = j - 1
a[i], a[j] = a[j], a[i] # swap the values
r = a[i+1:last]
r.reverse()
a[i+1:last] = r
yield list(a)
break
if i == first:
a.reverse()
return
if __name__ == '__main__':
for n in range(5):
for a in permute_in_place(range(1, n+1)):
print a
print
for a in permute_in_place([0, 0, 1, 1, 1]):
print a
print
这是初始排序后生成排列的渐近最优方式O(n*n!)。
有n个!最多进行一次置换,且具有下一次置换(..),以O(n)时间复杂度运行
在3个步骤中,
找到最大的j,使a[j]可以增加以最小可行量增加a[j]找到扩展新a[0..j]的字典最少方法
'''
Lexicographic permutation generation
consider example array state of [1,5,6,4,3,2] for sorted [1,2,3,4,5,6]
after 56432(treat as number) ->nothing larger than 6432(using 6,4,3,2) beginning with 5
so 6 is next larger and 2345(least using numbers other than 6)
so [1, 6,2,3,4,5]
'''
def hasNextPermutation(array, len):
' Base Condition '
if(len ==1):
return False
'''
Set j = last-2 and find first j such that a[j] < a[j+1]
If no such j(j==-1) then we have visited all permutations
after this step a[j+1]>=..>=a[len-1] and a[j]<a[j+1]
a[j]=5 or j=1, 6>5>4>3>2
'''
j = len -2
while (j >= 0 and array[j] >= array[j + 1]):
j= j-1
if(j==-1):
return False
# print(f"After step 2 for j {j} {array}")
'''
decrease l (from n-1 to j) repeatedly until a[j]<a[l]
Then swap a[j], a[l]
a[l] is the smallest element > a[j] that can follow a[l]...a[j-1] in permutation
before swap we have a[j+1]>=..>=a[l-1]>=a[l]>a[j]>=a[l+1]>=..>=a[len-1]
after swap -> a[j+1]>=..>=a[l-1]>=a[j]>a[l]>=a[l+1]>=..>=a[len-1]
a[l]=6 or l=2, j=1 just before swap [1, 5, 6, 4, 3, 2]
after swap [1, 6, 5, 4, 3, 2] a[l]=5, a[j]=6
'''
l = len -1
while(array[j] >= array[l]):
l = l-1
# print(f"After step 3 for l={l}, j={j} before swap {array}")
array[j], array[l] = array[l], array[j]
# print(f"After step 3 for l={l} j={j} after swap {array}")
'''
Reverse a[j+1...len-1](both inclusive)
after reversing [1, 6, 2, 3, 4, 5]
'''
array[j+1:len] = reversed(array[j+1:len])
# print(f"After step 4 reversing {array}")
return True
array = [1,2,4,4,5]
array.sort()
len = len(array)
count =1
print(array)
'''
The algorithm visits every permutation in lexicographic order
generating one by one
'''
while(hasNextPermutation(array, len)):
print(array)
count = count +1
# The number of permutations will be n! if no duplicates are present, else less than that
# [1,4,3,3,2] -> 5!/2!=60
print(f"Number of permutations: {count}")
常规实现(无收益-将在内存中完成所有操作):
def getPermutations(array):
if len(array) == 1:
return [array]
permutations = []
for i in range(len(array)):
# get all perm's of subarray w/o current item
perms = getPermutations(array[:i] + array[i+1:])
for p in perms:
permutations.append([array[i], *p])
return permutations
产量实施:
def getPermutations(array):
if len(array) == 1:
yield array
else:
for i in range(len(array)):
perms = getPermutations(array[:i] + array[i+1:])
for p in perms:
yield [array[i], *p]
基本思想是在第一个位置遍历数组中的所有元素,然后在第二个位置遍历所有其他元素,而没有为第一个位置选择的元素,等等。您可以使用递归来实现这一点,其中停止条件是到达一个由1个元素组成的数组,在这种情况下,您返回该数组。
用递归求解,遍历元素,取第i个元素,然后问自己:“其余项目的排列是什么”,直到没有更多的元素。
我在这里解释了解决方案:https://www.youtube.com/watch?v=_7GE7psS2b4
class Solution:
def permute(self,nums:List[int])->List[List[int]]:
res=[]
def dfs(nums,path):
if len(nums)==0:
res.append(path)
for i in range(len(nums)):
dfs(nums[:i]+nums[i+1:],path+[nums[i]])
dfs(nums,[])
return res