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
如果有人喜欢这个丑陋的单行线(虽然只适用于字符串):
def p(a):
return a if len(a) == 1 else [[a[i], *j] for i in range(len(a)) for j in p(a[:i] + a[i + 1:])]
其他回答
递归之美:
>>> import copy
>>> def perm(prefix,rest):
... for e in rest:
... new_rest=copy.copy(rest)
... new_prefix=copy.copy(prefix)
... new_prefix.append(e)
... new_rest.remove(e)
... if len(new_rest) == 0:
... print new_prefix + new_rest
... continue
... perm(new_prefix,new_rest)
...
>>> perm([],['a','b','c','d'])
['a', 'b', 'c', 'd']
['a', 'b', 'd', 'c']
['a', 'c', 'b', 'd']
['a', 'c', 'd', 'b']
['a', 'd', 'b', 'c']
['a', 'd', 'c', 'b']
['b', 'a', 'c', 'd']
['b', 'a', 'd', 'c']
['b', 'c', 'a', 'd']
['b', 'c', 'd', 'a']
['b', 'd', 'a', 'c']
['b', 'd', 'c', 'a']
['c', 'a', 'b', 'd']
['c', 'a', 'd', 'b']
['c', 'b', 'a', 'd']
['c', 'b', 'd', 'a']
['c', 'd', 'a', 'b']
['c', 'd', 'b', 'a']
['d', 'a', 'b', 'c']
['d', 'a', 'c', 'b']
['d', 'b', 'a', 'c']
['d', 'b', 'c', 'a']
['d', 'c', 'a', 'b']
['d', 'c', 'b', 'a']
人们确实可以对每个排列的第一个元素进行迭代,正如tzwen的答案。然而,这样编写此解决方案更有效:
def all_perms(elements):
if len(elements) <= 1:
yield elements # Only permutation possible = no permutation
else:
# Iteration over the first element in the result permutation:
for (index, first_elmt) in enumerate(elements):
other_elmts = elements[:index]+elements[index+1:]
for permutation in all_perms(other_elmts):
yield [first_elmt] + permutation
这个解决方案大约快了30%,显然是因为递归以len(元素)<=1而不是0结尾。它的内存效率也高得多,因为它使用了一个生成器函数(通过yield),就像Riccardo Reyes的解决方案一样。
对于性能,一个由Knuth启发的numpy解决方案(第22页):
from numpy import empty, uint8
from math import factorial
def perms(n):
f = 1
p = empty((2*n-1, factorial(n)), uint8)
for i in range(n):
p[i, :f] = i
p[i+1:2*i+1, :f] = p[:i, :f] # constitution de blocs
for j in range(i):
p[:i+1, f*(j+1):f*(j+2)] = p[j+1:j+i+2, :f] # copie de blocs
f = f*(i+1)
return p[:n, :]
复制大量内存可节省时间-它比列表(itertools.permutations(range(n))快20倍:
In [1]: %timeit -n10 list(permutations(range(10)))
10 loops, best of 3: 815 ms per loop
In [2]: %timeit -n100 perms(10)
100 loops, best of 3: 40 ms per loop
我使用了一种基于阶乘数系统的算法——对于长度为n的列表,您可以逐项组装每个排列,从每个阶段留下的项目中进行选择。第一项有n个选项,第二项有n-1个选项,最后一项只有一个选项,因此可以使用阶乘数系统中数字的数字作为索引。这是数字0到n-1对应于词典顺序中的所有可能的排列。
from math import factorial
def permutations(l):
permutations=[]
length=len(l)
for x in xrange(factorial(length)):
available=list(l)
newPermutation=[]
for radix in xrange(length, 0, -1):
placeValue=factorial(radix-1)
index=x/placeValue
newPermutation.append(available.pop(index))
x-=index*placeValue
permutations.append(newPermutation)
return permutations
permutations(range(3))
输出:
[[0, 1, 2], [0, 2, 1], [1, 0, 2], [1, 2, 0], [2, 0, 1], [2, 1, 0]]
此方法是非递归的,但在我的计算机上速度稍慢,xrange在n!太大,无法转换为C长整数(我的n=13)。当我需要它的时候,它已经足够了,但它远没有itertools.permutations。
def permutation(word, first_char=None):
if word == None or len(word) == 0: return []
if len(word) == 1: return [word]
result = []
first_char = word[0]
for sub_word in permutation(word[1:], first_char):
result += insert(first_char, sub_word)
return sorted(result)
def insert(ch, sub_word):
arr = [ch + sub_word]
for i in range(len(sub_word)):
arr.append(sub_word[i:] + ch + sub_word[:i])
return arr
assert permutation(None) == []
assert permutation('') == []
assert permutation('1') == ['1']
assert permutation('12') == ['12', '21']
print permutation('abc')
输出:['abc','acb','bac','bca','cab','cba']