将列表划分为的最佳方式是什么 大致 等份?例如,如果列表有 7 个元素,并将其分为 2 部分,我们希望在一部分中获取 3 个元素,而另一部分应该有 4 个元素。

我正在寻找类似的东西 even_split(L, n) 那打破了 L 进入 n 部分。

def chunks(L, n):
    """ Yield successive n-sized chunks from L.
    """
    for i in xrange(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(迭代它并获取每列的第一个元素,将该元素称为第一部分,然后获取第二个元素并将其放入第二部分,等等),但这会破坏项目的顺序。

有帮助吗?

解决方案

下面是一个可以工作:

def chunkIt(seq, num):
    avg = len(seq) / float(num)
    out = []
    last = 0.0

    while last < len(seq):
        out.append(seq[int(last):int(last + avg)])
        last += avg

    return out

测试:

>>> chunkIt(range(10), 3)
[[0, 1, 2], [3, 4, 5], [6, 7, 8, 9]]
>>> chunkIt(range(11), 3)
[[0, 1, 2], [3, 4, 5, 6], [7, 8, 9, 10]]
>>> chunkIt(range(12), 3)
[[0, 1, 2, 3], [4, 5, 6, 7], [8, 9, 10, 11]]

其他提示

您可以编写它相当简单的列表生成:

def split(a, n):
    k, m = divmod(len(a), n)
    return (a[i * k + min(i, m):(i + 1) * k + min(i + 1, m)] for i in xrange(n))

示例:

>>> list(split(range(11), 3))
[[0, 1, 2, 3], [4, 5, 6, 7], [8, 9, 10]]

只要你不想做傻事像连续块:

>>> def chunkify(lst,n):
...     return [lst[i::n] for i in xrange(n)]
... 
>>> chunkify(range(13), 3)
[[0, 3, 6, 9, 12], [1, 4, 7, 10], [2, 5, 8, 11]]

这是 存在的理由 为了 numpy.array_split*:

>>> L
[0, 1, 2, 3, 4, 5, 6, 7]
>>> print(*np.array_split(L, 3))
[0 1 2] [3 4 5] [6 7]
>>> print(*np.array_split(range(10), 4))
[0 1 2] [3 4 5] [6 7] [8 9]

*归功于 比雷埃夫斯零号 在6号房间

的代码变更为屈服n块而不是n的块:

def chunks(l, n):
    """ Yield n successive chunks from l.
    """
    newn = int(len(l) / n)
    for i in xrange(0, n-1):
        yield l[i*newn:i*newn+newn]
    yield l[n*newn-newn:]

l = range(56)
three_chunks = chunks (l, 3)
print three_chunks.next()
print three_chunks.next()
print three_chunks.next()

其给出:

[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]

这将分配额外元件,其并不完美,但在您的“大致N个相等的部分”规范以及最后一组:-)通过的是,我的意思是56层的元件将是更好为(19,19,18)而这给出(18,18,20)。

您可以得到下面的代码更加平衡输出:

#!/usr/bin/python
def chunks(l, n):
    """ Yield n successive chunks from l.
    """
    newn = int(1.0 * len(l) / n + 0.5)
    for i in xrange(0, n-1):
        yield l[i*newn:i*newn+newn]
    yield l[n*newn-newn:]

l = range(56)
three_chunks = chunks (l, 3)
print three_chunks.next()
print three_chunks.next()
print three_chunks.next()

其中输出:

[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]

如果分割n元件成大致k块可以使n % k块1个元件比另一组块更大分发额外的元素。

下面的代码给你的长度为组块:

[(n // k) + (1 if i < (n % k) else 0) for i in range(k)]

实施例:n=11, k=3结果[4, 4, 3]

可以然后很容易地计算起始indizes为组块:

[i * (n // k) + min(i, n % k) for i in range(k)]

实施例:n=11, k=3结果[0, 4, 8]

使用i+1th块为界,我们得到的名单i与LEN l

nth块
l[i * (n // k) + min(i, n % k):(i+1) * (n // k) + min(i+1, n % k)]

作为最后的步骤中,使用从所有的块创建列表列表理解:

[l[i * (n // k) + min(i, n % k):(i+1) * (n // k) + min(i+1, n % k)] for i in range(k)]

实施例:n=11, k=3, l=range(n)结果[range(0, 4), range(4, 8), range(8, 11)]

下面是一种增加了None使列表相等长度

>>> from itertools import izip_longest
>>> def chunks(l, n):
    """ Yield n successive chunks from l. Pads extra spaces with None
    """
    return list(zip(*izip_longest(*[iter(l)]*n)))

>>> l=range(54)

>>> chunks(l,3)
[(0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45, 48, 51), (1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 34, 37, 40, 43, 46, 49, 52), (2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 32, 35, 38, 41, 44, 47, 50, 53)]

>>> chunks(l,4)
[(0, 4, 8, 12, 16, 20, 24, 28, 32, 36, 40, 44, 48, 52), (1, 5, 9, 13, 17, 21, 25, 29, 33, 37, 41, 45, 49, 53), (2, 6, 10, 14, 18, 22, 26, 30, 34, 38, 42, 46, 50, None), (3, 7, 11, 15, 19, 23, 27, 31, 35, 39, 43, 47, 51, None)]

>>> chunks(l,5)
[(0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50), (1, 6, 11, 16, 21, 26, 31, 36, 41, 46, 51), (2, 7, 12, 17, 22, 27, 32, 37, 42, 47, 52), (3, 8, 13, 18, 23, 28, 33, 38, 43, 48, 53), (4, 9, 14, 19, 24, 29, 34, 39, 44, 49, None)]

这将通过单个表达做分割:

>>> myList = range(18)
>>> parts = 5
>>> [myList[(i*len(myList))//parts:((i+1)*len(myList))//parts] for i in range(parts)]
[[0, 1, 2], [3, 4, 5, 6], [7, 8, 9], [10, 11, 12, 13], [14, 15, 16, 17]]

在这个例子中的列表具有18的尺寸和被分成5份。的部件不同的在不超过一个元件的尺寸。

请参阅 more_itertools.divide

n = 2

[list(x) for x in mit.divide(n, range(5, 11))]
# [[5, 6, 7], [8, 9, 10]]

[list(x) for x in mit.divide(n, range(5, 12))]
# [[5, 6, 7, 8], [9, 10, 11]]

安装经由 > pip install more_itertools

看一看 numpy.split

>>> a = numpy.array([1,2,3,4])
>>> numpy.split(a, 2)
[array([1, 2]), array([3, 4])]

使用numpy.linspace方法实现。

只要指定要在阵列在to.The区划划分将是几乎相等的大小的部件的数量。

实施例:     

import numpy as np   
a=np.arange(10)
print "Input array:",a 
parts=3
i=np.linspace(np.min(a),np.max(a)+1,parts+1)
i=np.array(i,dtype='uint16') # Indices should be floats
split_arr=[]
for ind in range(i.size-1):
    split_arr.append(a[i[ind]:i[ind+1]]
print "Array split in to %d parts : "%(parts),split_arr

给出:

Input array: [0 1 2 3 4 5 6 7 8 9]
Array split in to 3 parts :  [array([0, 1, 2]), array([3, 4, 5]), array([6, 7, 8, 9])]

下面是我的解决方案:

def chunks(l, amount):
    if amount < 1:
        raise ValueError('amount must be positive integer')
    chunk_len = len(l) // amount
    leap_parts = len(l) % amount
    remainder = amount // 2  # make it symmetrical
    i = 0
    while i < len(l):
        remainder += leap_parts
        end_index = i + chunk_len
        if remainder >= amount:
            remainder -= amount
            end_index += 1
        yield l[i:end_index]
        i = end_index

可生产

    >>> list(chunks([1, 2, 3, 4, 5, 6, 7], 3))
    [[1, 2], [3, 4, 5], [6, 7]]

下面是一个发生器,它可处理块的任何正(整数)编号。如果块的数量比输入列表长度更大一些大块是空的。短的和长的块之间的这种算法交替,而不是分离它们。

我还包含一些代码用于测试ragged_chunks功能。

''' 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')

我们可以让这个的的通过导出倍增到range调用更有效,但我觉得以前的版本更易读(和烘干机)。

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

使用列表理解:

def divide_list_to_chunks(list_, n):
    return [list_[start::n] for start in range(n)]

我的解决办法,很容易理解

def split_list(lst, n):
    splitted = []
    for i in reversed(range(1, n + 1)):
        split_point = len(lst)//i
        splitted.append(lst[:split_point])
        lst = lst[split_point:]
    return splitted

和此页面上最短的一行(通过我的女孩写的)

def split(l, n):
    return [l[int(i*len(l)/n):int((i+1)*len(l)/n-1)] for i in range(n)]

说你要分成5个部分:

p1, p2, p3, p4, p5 = np.split(df, 5)

另一种方法是这样的,这里的想法是使用石斑鱼,但摆脱None的。在这种情况下,我们就必须从列表中后期在列表的第一部分的元素,和“larger_parts”形成的全部“small_parts”。的“较大的部件”长度LEN(small_parts)+ 1我们需要考虑x作为两个不同的子部分。

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

我有它设置返回的方式元组的列表:

>>> 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)]
>>> 

下面是另一种变型的不同点差“剩余”的元素的所有块之间平均,一次一个,直到有没有左边。在该实现中,将发生更大的块在开始该过程。

def chunks(l, k):
  """ Yield k successive chunks from l."""
  if k < 1:
    yield []
    raise StopIteration
  n = len(l)
  avg = n/k
  remainders = n % k
  start, end = 0, avg
  while start < n:
    if remainders > 0:
      end = end + 1
      remainders = remainders - 1
    yield l[start:end]
    start, end = end, end+avg

例如,从14个元件的列表生成4组块:

>>> list(chunks(range(14), 4))
[[0, 1, 2, 3], [4, 5, 6, 7], [8, 9, 10], [11, 12, 13]]
>>> map(len, list(chunks(range(14), 4)))
[4, 4, 3, 3]

同为工作的答案,但考虑到与列表尺寸比chuncks的数量。

def chunkify(lst,n):
    [ lst[i::n] for i in xrange(n if n < len(lst) else len(lst)) ]

如果n(块的数量)是7和LST(列表以除)为[1,2,3]的块是[[0],[1],[2]]而非[[0] [1],[2],[],[],[],[]]

您也可以使用:

split=lambda x,n: x if not x else [x[:n]]+[split([] if not -(len(x)-n) else x[-(len(x)-n):],n)][0]

split([1,2,3,4,5,6,7,8,9],2)

[[1, 2], [3, 4], [5, 6], [7, 8], [9]]
#!/usr/bin/python


first_names = ['Steve', 'Jane', 'Sara', 'Mary','Jack','Bob', 'Bily', 'Boni', 'Chris','Sori', 'Will', 'Won','Li']

def chunks(l, n):
for i in range(0, len(l), n):
    # Create an index range for l of n items:
    yield l[i:i+n]

result = list(chunks(first_names, 5))
print result

链接摘下来的,这是什么帮助了我。我有一个预定义列表。

我在这种情况下写的代码自己:

def chunk_ports(port_start, port_end, portions):
    if port_end < port_start:
        return None

    total = port_end - port_start + 1

    fractions = int(math.floor(float(total) / portions))

    results = []

    # No enough to chuck.
    if fractions < 1:
        return None

    # Reverse, so any additional items would be in the first range.
    _e = port_end
    for i in range(portions, 0, -1):
        print "i", i

        if i == 1:
            _s = port_start
        else:
            _s = _e - fractions + 1

        results.append((_s, _e))

        _e = _s - 1

    results.reverse()

    return results

divide_ports(1,10,9)将返回

[(1, 2), (3, 3), (4, 4), (5, 5), (6, 6), (7, 7), (8, 8), (9, 9), (10, 10)]

此代码对我的作品(Python3兼容):

def chunkify(tab, num):
    return [tab[i*num: i*num+num] for i in range(len(tab)//num+(1 if len(tab)%num else 0))]

示例(对于字节组型,但它工作于的列表 S作为孔):

b = bytearray(b'\x01\x02\x03\x04\x05\x06\x07\x08')
>>> chunkify(b,3)
[bytearray(b'\x01\x02\x03'), bytearray(b'\x04\x05\x06'), bytearray(b'\x07\x08')]
>>> chunkify(b,4)
[bytearray(b'\x01\x02\x03\x04'), bytearray(b'\x05\x06\x07\x08')]

此提供一个长度的数据块<= N,> = 0

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]]

我试图解决方案的大部分,但他们没有工作,我的情况,所以我提出一个新的功能,对于大多数情况下和任何类型的数组的工作:

import math

def chunkIt(seq, num):
    seqLen = len(seq)
    total_chunks = math.ceil(seqLen / num)
    items_per_chunk = num
    out = []
    last = 0

    while last < seqLen:
        out.append(seq[last:(last + items_per_chunk)])
        last += items_per_chunk

    return out
def evenly(l, n):
    len_ = len(l)
    split_size = len_ // n
    split_size = n if not split_size else split_size
    offsets = [i for i in range(0, len_, split_size)]
    return [l[offset:offset + split_size] for offset in offsets]

示例:

l = [a for a in range(97)]应包括10重量份,各自具有除了最后一个9个元素。

<强>输出:

[[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]]

四舍五入的linspace并用它作为一个指标大于什么amit12690提出了一种简单的解决方案。

function chunks=chunkit(array,num)

index = round(linspace(0,size(array,2),num+1));

chunks = cell(1,num);

for x = 1:num
chunks{x} = array(:,index(x)+1:index(x+1));
end
end
许可以下: CC-BY-SA归因
不隶属于 StackOverflow
scroll top