将清单大致等份的最佳方法是什么?例如，如果列表有7个元素，并将其分为2部分，我们希望其中一部分有3个元素，而另一部分应该有4个元素。
我正在寻找类似even_split(L, n)的东西，它将L分解为n部分。
def chunks(L, n):
""" Yield successive n-sized chunks from L.
"""
for i in range(0, len(L), n):
yield L[i:i+n]
上面的代码给出了3个块，而不是3个块。我可以简单地转置(遍历这个，取每列的第一个元素，称之为第一部分，然后取第二个元素，把它放在第二部分，等等)，但这破坏了项目的顺序。
当前回答
另一种方法是这样的，这里的想法是用石斑鱼，但要去掉None。在本例中，所有的'small_parts'都由列表第一部分的元素组成，'larger_parts'则由列表的后一部分组成。' bigger parts'的长度为len(small_parts) + 1。我们需要把x看成两个不同的子部分。
from itertools import izip_longest
import numpy as np
def grouper(n, iterable, fillvalue=None): # This is grouper from itertools
"grouper(3, 'ABCDEFG', 'x') --> ABC DEF Gxx"
args = [iter(iterable)] * n
return izip_longest(fillvalue=fillvalue, *args)
def another_chunk(x,num):
extra_ele = len(x)%num #gives number of parts that will have an extra element
small_part = int(np.floor(len(x)/num)) #gives number of elements in a small part
new_x = list(grouper(small_part,x[:small_part*(num-extra_ele)]))
new_x.extend(list(grouper(small_part+1,x[small_part*(num-extra_ele):])))
return new_x
我设置它的方式返回一个元组列表:
>>> x = range(14)
>>> another_chunk(x,3)
[(0, 1, 2, 3), (4, 5, 6, 7, 8), (9, 10, 11, 12, 13)]
>>> another_chunk(x,4)
[(0, 1, 2), (3, 4, 5), (6, 7, 8, 9), (10, 11, 12, 13)]
>>> another_chunk(x,5)
[(0, 1), (2, 3, 4), (5, 6, 7), (8, 9, 10), (11, 12, 13)]
>>>
其他回答
这里有一个单独的函数，它处理了大多数不同的分裂情况:
def splitList(lst, into):
'''Split a list into parts.
:Parameters:
into (str) = Split the list into parts defined by the following:
'<n>parts' - Split the list into n parts.
ex. 2 returns: [[1, 2, 3, 5], [7, 8, 9]] from [1,2,3,5,7,8,9]
'<n>parts+' - Split the list into n equal parts with any trailing remainder.
ex. 2 returns: [[1, 2, 3], [5, 7, 8], [9]] from [1,2,3,5,7,8,9]
'<n>chunks' - Split into sublists of n size.
ex. 2 returns: [[1,2], [3,5], [7,8], [9]] from [1,2,3,5,7,8,9]
'contiguous' - The list will be split by contiguous numerical values.
ex. 'contiguous' returns: [[1,2,3], [5], [7,8,9]] from [1,2,3,5,7,8,9]
'range' - The values of 'contiguous' will be limited to the high and low end of each range.
ex. 'range' returns: [[1,3], [5], [7,9]] from [1,2,3,5,7,8,9]
:Return:
(list)
'''
from string import digits, ascii_letters, punctuation
mode = into.lower().lstrip(digits)
digit = into.strip(ascii_letters+punctuation)
n = int(digit) if digit else None
if n:
if mode=='parts':
n = len(lst)*-1 // n*-1 #ceil
elif mode=='parts+':
n = len(lst) // n
return [lst[i:i+n] for i in range(0, len(lst), n)]
elif mode=='contiguous' or mode=='range':
from itertools import groupby
from operator import itemgetter
try:
contiguous = [list(map(itemgetter(1), g)) for k, g in groupby(enumerate(lst), lambda x: int(x[0])-int(x[1]))]
except ValueError as error:
print ('{} in splitList\n # Error: {} #\n {}'.format(__file__, error, lst))
return lst
if mode=='range':
return [[i[0], i[-1]] if len(i)>1 else (i) for i in contiguous]
return contiguous
r = splitList([1, '2', 3, 5, '7', 8, 9], into='2parts')
print (r) #returns: [[1, '2', 3, 5], ['7', 8, 9]]
假设你想把一个列表[1、2、3、4、5、6、7、8]分成3个元素列表
如[[1,2,3]，[4,5,6]，[7,8]]，如果剩下的最后一个元素小于3，则将它们分组在一起。
my_list = [1, 2, 3, 4, 5, 6, 7, 8]
my_list2 = [my_list[i:i+3] for i in range(0, len(my_list), 3)]
print(my_list2)
输出:[[1,2,3]，[4,5,6]，[7,8]]
其中一部分的长度为3。用你自己的块大小替换3。
另一个简单可读的chunker的尝试。
def chunk(iterable, count): # returns a *generator* that divides `iterable` into `count` of contiguous chunks of similar size
assert count >= 1
return (iterable[int(_*len(iterable)/count+0.5):int((_+1)*len(iterable)/count+0.5)] for _ in range(count))
print("Chunk count: ", len(list( chunk(range(105),10))))
print("Chunks: ", list( chunk(range(105),10)))
print("Chunks: ", list(map(list,chunk(range(105),10))))
print("Chunk lengths:", list(map(len, chunk(range(105),10))))
print("Testing...")
for iterable_length in range(100):
for chunk_count in range(1,100):
chunks = list(chunk(range(iterable_length),chunk_count))
assert chunk_count == len(chunks)
assert iterable_length == sum(map(len,chunks))
assert all(map(lambda _:abs(len(_)-iterable_length/chunk_count)<=1,chunks))
print("Okay")
输出:
Chunk count: 10
Chunks: [range(0, 11), range(11, 21), range(21, 32), range(32, 42), range(42, 53), range(53, 63), range(63, 74), range(74, 84), range(84, 95), range(95, 105)]
Chunks: [[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10], [11, 12, 13, 14, 15, 16, 17, 18, 19, 20], [21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31], [32, 33, 34, 35, 36, 37, 38, 39, 40, 41], [42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52], [53, 54, 55, 56, 57, 58, 59, 60, 61, 62], [63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73], [74, 75, 76, 77, 78, 79, 80, 81, 82, 83], [84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94], [95, 96, 97, 98, 99, 100, 101, 102, 103, 104]]
Chunk lengths: [11, 10, 11, 10, 11, 10, 11, 10, 11, 10]
Testing...
Okay
它提供长度<= n， >= 0的块
def
chunkify(lst, n):
num_chunks = int(math.ceil(len(lst) / float(n))) if n < len(lst) else 1
return [lst[n*i:n*(i+1)] for i in range(num_chunks)]
例如
>>> chunkify(range(11), 3)
[[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 10]]
>>> chunkify(range(11), 8)
[[0, 1, 2, 3, 4, 5, 6, 7], [8, 9, 10]]
这里有一个生成器，可以处理任何正(整数)数量的块。如果块的数量大于输入列表的长度，一些块将为空。该算法在短块和长块之间交替，而不是将它们分开。
我还包含了一些用于测试ragged_chunks函数的代码。
''' Split a list into "ragged" chunks
The size of each chunk is either the floor or ceiling of len(seq) / chunks
chunks can be > len(seq), in which case there will be empty chunks
Written by PM 2Ring 2017.03.30
'''
def ragged_chunks(seq, chunks):
size = len(seq)
start = 0
for i in range(1, chunks + 1):
stop = i * size // chunks
yield seq[start:stop]
start = stop
# test
def test_ragged_chunks(maxsize):
for size in range(0, maxsize):
seq = list(range(size))
for chunks in range(1, size + 1):
minwidth = size // chunks
#ceiling division
maxwidth = -(-size // chunks)
a = list(ragged_chunks(seq, chunks))
sizes = [len(u) for u in a]
deltas = all(minwidth <= u <= maxwidth for u in sizes)
assert all((sum(a, []) == seq, sum(sizes) == size, deltas))
return True
if test_ragged_chunks(100):
print('ok')
我们可以通过将乘法导出到range调用中来稍微提高效率，但我认为以前的版本更易于阅读(和dry)。
def ragged_chunks(seq, chunks):
size = len(seq)
start = 0
for i in range(size, size * chunks + 1, size):
stop = i // chunks
yield seq[start:stop]
start = stop
