How do I make a flat list out of a list of lists?

Given a list of lists l,

flat_list = [item for sublist in l for item in sublist]

which means:

flat_list = []
for sublist in l:
    for item in sublist:
        flat_list.append(item)

is faster than the shortcuts posted so far. (l is the list to flatten.)

Here is the corresponding function:

def flatten(l):
    return [item for sublist in l for item in sublist]

As evidence, you can use the timeit module in the standard library:

$ python -mtimeit -s'l=[[1,2,3],[4,5,6], [7], [8,9]]*99' '[item for sublist in l for item in sublist]'
10000 loops, best of 3: 143 usec per loop
$ python -mtimeit -s'l=[[1,2,3],[4,5,6], [7], [8,9]]*99' 'sum(l, [])'
1000 loops, best of 3: 969 usec per loop
$ python -mtimeit -s'l=[[1,2,3],[4,5,6], [7], [8,9]]*99' 'reduce(lambda x,y: x+y,l)'
1000 loops, best of 3: 1.1 msec per loop

Explanation: the shortcuts based on + (including the implied use in sum) are, of necessity, O(L**2) when there are L sublists -- as the intermediate result list keeps getting longer, at each step a new intermediate result list object gets allocated, and all the items in the previous intermediate result must be copied over (as well as a few new ones added at the end). So, for simplicity and without actual loss of generality, say you have L sublists of I items each: the first I items are copied back and forth L-1 times, the second I items L-2 times, and so on; total number of copies is I times the sum of x for x from 1 to L excluded, i.e., I * (L**2)/2.

The list comprehension just generates one list, once, and copies each item over (from its original place of residence to the result list) also exactly once.

You can use itertools.chain():

>>> import itertools
>>> list2d = [[1,2,3], [4,5,6], [7], [8,9]]
>>> merged = list(itertools.chain(*list2d))

Or you can use itertools.chain.from_iterable() which doesn't require unpacking the list with the * operator:

>>> import itertools
>>> list2d = [[1,2,3], [4,5,6], [7], [8,9]]
>>> merged = list(itertools.chain.from_iterable(list2d))

This approach is arguably more readable than [item for sublist in l for item in sublist] and appears to be faster too:

$ python3 -mtimeit -s'l=[[1,2,3],[4,5,6], [7], [8,9]]*99;import itertools' 'list(itertools.chain.from_iterable(l))'
20000 loops, best of 5: 10.8 usec per loop
$ python3 -mtimeit -s'l=[[1,2,3],[4,5,6], [7], [8,9]]*99' '[item for sublist in l for item in sublist]'
10000 loops, best of 5: 21.7 usec per loop
$ python3 -mtimeit -s'l=[[1,2,3],[4,5,6], [7], [8,9]]*99' 'sum(l, [])'
1000 loops, best of 5: 258 usec per loop
$ python3 -mtimeit -s'l=[[1,2,3],[4,5,6], [7], [8,9]]*99;from functools import reduce' 'reduce(lambda x,y: x+y,l)'
1000 loops, best of 5: 292 usec per loop
$ python3 --version
Python 3.7.5rc1

Note from the author: This is very inefficient. But fun, because monoids are awesome.

>>> xss = [[1, 2, 3], [4, 5, 6], [7], [8, 9]]
>>> sum(xss, [])
[1, 2, 3, 4, 5, 6, 7, 8, 9]

sum sums the elements of the iterable xss, and uses the second argument as the initial value [] for the sum. (The default initial value is 0, which is not a list.)

Because you are summing nested lists, you actually get [1,3]+[2,4] as a result of sum([[1,3],[2,4]],[]), which is equal to [1,3,2,4].

Note that only works on lists of lists. For lists of lists of lists, you'll need another solution.

Using functools.reduce, which adds an accumulated list xs to the next list ys:

from functools import reduce
xss = [[1,2,3], [4,5,6], [7], [8,9]]
out = reduce(lambda xs, ys: xs + ys, xss)

Output:

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

A faster way using operator.concat:

from functools import reduce
import operator
xss = [[1,2,3], [4,5,6], [7], [8,9]]
out = reduce(operator.concat, xss)

Output:

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

The reason your function didn't work is because the extend extends an array in-place and doesn't return it. You can still return x from lambda, using something like this:

reduce(lambda x,y: x.extend(y) or x, l)

Note: extend is more efficient than + on lists.

matplotlib.cbook.flatten() will work for nested lists even if they nest more deeply than the example.

import matplotlib
l = [[1, 2, 3], [4, 5, 6], [7], [8, 9]]
print(list(matplotlib.cbook.flatten(l)))
l2 = [[1, 2, 3], [4, 5, 6], [7], [8, [9, 10, [11, 12, [13]]]]]
print(list(matplotlib.cbook.flatten(l2)))

Result:

[1, 2, 3, 4, 5, 6, 7, 8, 9]
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]

This is 18x faster than underscore._.flatten:

Average time over 1000 trials of matplotlib.cbook.flatten: 2.55e-05 sec
Average time over 1000 trials of underscore._.flatten: 4.63e-04 sec
(time for underscore._)/(time for matplotlib.cbook) = 18.1233394636

According your list [[1, 2, 3], [4, 5, 6], [7], [8, 9]] which is 1 list level, we can simply use sum(list,[]) without using any libraries

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

To extend the advantage of this method when there is a tuple or number existing inside. Simply adding a mapping function for each element by map to the list

#For only tuple
sum(list(map(list,[[1, 2, 3], (4, 5, 6), (7,), [8, 9]])),[])
# [1, 2, 3, 4, 5, 6, 7, 8, 9]

#In general

def convert(x):
    if type(x) is int or type(x) is float:
           return [x]
    else:
           return list(x)

sum(list(map(convert,[[1, 2, 3], (4, 5, 6), 7, [8, 9]])),[])
# [1, 2, 3, 4, 5, 6, 7, 8, 9]

In here, there is a clear explanation of the drawback in terms of memory for this approach. In short, it recursively creates list objects, which should be avoided :(

You can use the list extend method. It shows to be the fastest:

flat_list = []
for sublist in l:
    flat_list.extend(sublist)

Performance:

import functools
import itertools
import numpy
import operator
import perfplot


def functools_reduce_iconcat(a):
    return functools.reduce(operator.iconcat, a, [])


def itertools_chain(a):
    return list(itertools.chain.from_iterable(a))


def numpy_flat(a):
    return list(numpy.array(a).flat)


def extend(a):
    n = []

    list(map(n.extend, a))

    return n


perfplot.show(
    setup = lambda n: [list(range(10))] * n,
    kernels = [
        functools_reduce_iconcat, extend, itertools_chain, numpy_flat
        ],
    n_range = [2**k for k in range(16)],
    xlabel = 'num lists',
    )

Output:

Use two for in list comprehension:

l = [[1, 2, 3], [4, 5, 6], [7], [8, 9]]
flat_l = [e for v in l for e in v]
print(flat_l)

There are several answers with the same recursive appending scheme as below, but none makes use of try, which makes the solution more robust and Pythonic.

def flatten(itr):
    for x in itr:
        try:
            yield from flatten(x)
        except TypeError:
            yield x

Usage: this is a generator, and you typically want to enclose it in an iterable builder like list() or tuple() or use it in a for loop.

Advantages of this solution are:

• works with any kind of iterable (even future ones!)
• works with any combination and deepness of nesting
• works also if top level contains bare items
• no dependencies
• fast and efficient (you can flatten the nested iterable partially, without wasting time on the remaining part you don't need)
• versatile (you can use it to build an iterable of your choice or in a loop)

N.B.: Since all iterables are flattened, strings are decomposed into sequences of single characters. If you don't like/want such behavior, you can use the following version which filters out from flattening iterables like strings and bytes:

def flatten(itr):
    if type(itr) in (str,bytes):
        yield itr
    else:
        for x in itr:
            try:
                yield from flatten(x)
            except TypeError:
                yield x

If you are willing to give up a tiny amount of speed for a cleaner look, then you could use numpy.concatenate().tolist() or numpy.concatenate().ravel().tolist():

import numpy

l = [[1, 2, 3], [4, 5, 6], [7], [8, 9]] * 99

%timeit numpy.concatenate(l).ravel().tolist()
1000 loops, best of 3: 313 µs per loop

%timeit numpy.concatenate(l).tolist()
1000 loops, best of 3: 312 µs per loop

%timeit [item for sublist in l for item in sublist]
1000 loops, best of 3: 31.5 µs per loop

You can find out more here in the documentation, numpy.concatenate and numpy.ravel.

Note: Below applies to Python 3.3+ because it uses yield_from. six is also a third-party package, though it is stable. Alternately, you could use sys.version.

In the case of obj = [[1, 2,], [3, 4], [5, 6]], all of the solutions here are good, including list comprehension and itertools.chain.from_iterable.

However, consider this slightly more complex case:

>>> obj = [[1, 2, 3], [4, 5], 6, 'abc', [7], [8, [9, 10]]]

There are several problems here:

• One element, 6, is just a scalar; it's not iterable, so the above routes will fail here.
• One element, 'abc', is technically iterable (all strs are). However, reading between the lines a bit, you don't want to treat it as such--you want to treat it as a single element.
• The final element, [8, [9, 10]] is itself a nested iterable. Basic list comprehension and chain.from_iterable only extract "1 level down."

You can remedy this as follows:

>>> from collections import Iterable
>>> from six import string_types

>>> def flatten(obj):
...     for i in obj:
...         if isinstance(i, Iterable) and not isinstance(i, string_types):
...             yield from flatten(i)
...         else:
...             yield i


>>> list(flatten(obj))
[1, 2, 3, 4, 5, 6, 'abc', 7, 8, 9, 10]

Here, you check that the sub-element (1) is iterable with Iterable, an ABC from itertools, but also want to ensure that (2) the element is not "string-like."

This may not be the most efficient way, but I thought to put a one-liner (actually a two-liner). Both versions will work on arbitrary hierarchy nested lists, and exploits language features (Python 3.5) and recursion.

def make_list_flat (l):
    flist = []
    flist.extend ([l]) if (type (l) is not list) else [flist.extend (make_list_flat (e)) for e in l]
    return flist

a = [[1, 2], [[[[3, 4, 5], 6]]], 7, [8, [9, [10, 11], 12, [13, 14, [15, [[16, 17], 18]]]]]]
flist = make_list_flat(a)
print (flist)

The output is

[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18]

This works in a depth first manner. The recursion goes down until it finds a non-list element, then extends the local variable flist and then rolls back it to the parent. Whenever flist is returned, it is extended to the parent's flist in the list comprehension. Therefore, at the root, a flat list is returned.

The above one creates several local lists and returns them which are used to extend the parent's list. I think the way around for this may be creating a gloabl flist, like below.

a = [[1, 2], [[[[3, 4, 5], 6]]], 7, [8, [9, [10, 11], 12, [13, 14, [15, [[16, 17], 18]]]]]]
flist = []
def make_list_flat (l):
    flist.extend ([l]) if (type (l) is not list) else [make_list_flat (e) for e in l]

make_list_flat(a)
print (flist)

The output is again

[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18]

Although I am not sure at this time about the efficiency.

I wanted a solution which can deal with multiple nesting ([[1], [[[2]], [3]]], [1, 2, 3] for example), but would also not be recursive (I had a big level of recursion and I got a recursion error.

This is what I came up with:

def _flatten(l) -> Iterator[Any]:
    stack = l.copy()
    while stack:
        item = stack.pop()
        if isinstance(item, list):
            stack.extend(item)
        else:
            yield item


def flatten(l) -> Iterator[Any]:
    return reversed(list(_flatten(l)))

and tests:

@pytest.mark.parametrize('input_list, expected_output', [
    ([1, 2, 3], [1, 2, 3]),
    ([[1], 2, 3], [1, 2, 3]),
    ([[1], [2], 3], [1, 2, 3]),
    ([[1], [2], [3]], [1, 2, 3]),
    ([[1], [[2]], [3]], [1, 2, 3]),
    ([[1], [[[2]], [3]]], [1, 2, 3]),
])
def test_flatten(input_list, expected_output):
    assert list(flatten(input_list)) == expected_output

Not a one-liner, but seeing all the answers here, I guess this long list missed some pattern matching, so here it is :)

The two methods are probably not efficient, but anyway, it's easy to read (to me at least; perhaps I'm spoiled by functional programming):

def flat(x):
    match x:
        case []:
            return []
        case [[*sublist], *r]:
            return [*sublist, *flat(r)]

The second version considers lists of lists of lists... whatever the nesting:

def flat(x):
    match x:
        case []:
            return []
        case [[*sublist], *r]:
            return [*flat(sublist), *flat(r)]
        case [h, *r]:
            return [h, *flat(r)]

Another unusual approach that works for hetero- and homogeneous lists of integers:

from typing import List


def flatten(l: list) -> List[int]:
    """Flatten an arbitrary deep nested list of lists of integers.

    Examples:
        >>> flatten([1, 2, [1, [10]]])
        [1, 2, 1, 10]

    Args:
        l: Union[l, Union[int, List[int]]

    Returns:
        Flatted list of integer
    """
    return [int(i.strip('[ ]')) for i in str(l).split(',')]

A non-recursive function to flatten lists of lists of any depth:

def flatten_list(list1):
    out = []
    inside = list1
    while inside:
        x = inside.pop(0)
        if isinstance(x, list):
            inside[0:0] = x
        else:
            out.append(x)
    return out

l = [[[1,2],3,[4,[[5,6],7],[8]]],[9,10,11]]
flatten_list(l)
# [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]

If you want to unnest everything and keep a distinct list of elements, you could use this as well.

list_of_lists = [[1,2], [2,3], [3,4]]
list(set.union(*[set(s) for s in list_of_lists]))

For a list containing multiple list here a recursive solution that work for me and that i hope is correct:

# Question 4
def flatten(input_ls=[]) -> []:
    res_ls = []
    res_ls = flatten_recursive(input_ls, res_ls)

    print("Final flatten list solution is: \n", res_ls)

    return res_ls


def flatten_recursive(input_ls=[], res_ls=[]) -> []:
    tmp_ls = []

    for i in input_ls:
        if isinstance(i, int):
            res_ls.append(i)
        else:
            tmp_ls = i
            tmp_ls.append(flatten_recursive(i, res_ls))

    print(res_ls)
    return res_ls


flatten([0, 1, [2, 3], 4, [5, 6]])  # test
flatten([0, [[[1]]], [[2, 3], [4, [[5, 6]]]]])

Output:

[0, 1, 2, 3]
[0, 1, 2, 3, 4, 5, 6]
[0, 1, 2, 3, 4, 5, 6]
Final flatten list solution is: 
 [0, 1, 2, 3, 4, 5, 6]
[0, 1]
[0, 1]
[0, 1]
[0, 1, 2, 3]
[0, 1, 2, 3, 4, 5, 6]
[0, 1, 2, 3, 4, 5, 6]
[0, 1, 2, 3, 4, 5, 6]
[0, 1, 2, 3, 4, 5, 6]
[0, 1, 2, 3, 4, 5, 6]
Final flatten list solution is: 
 [0, 1, 2, 3, 4, 5, 6]

You can use the following:

def flatlst(lista):
    listaplana = []
    for k in lista: listaplana = listaplana + k
    return listaplana

If I want to add something to the great previous answers, here is my recursive flatten function which can flatten not only nested lists, but also any given container or any generally any object which can throw out items. This does also work for any depth of nesting and it is a lazy iterator which yields the items as requested:

def flatten(iterable):
    # These types won't considered a sequence or generally a container
    exclude = str, bytes

    for i in iterable:
        try:
            if isinstance(i, exclude):
                raise TypeError
            iter(i)
        except TypeError:
            yield i
        else:
            yield from flatten(i)

This way, you can exclude types you don't want to be flattened, like str or what else.

The idea is if an object can pass the iter() it's ready to yield items. So the iterable can have even generator expressions as an item.

Someone could argue: Why did you write this that generic when the OP didn't ask for it? OK, you're right. I just felt like this might help someone (like it did for myself).

Test cases:

lst1 = [1, {3}, (1, 6), [[3, 8]], [[[5]]], 9, ((((2,),),),)]
lst2 = ['3', B'A', [[[(i ** 2 for i in range(3))]]], range(3)]

print(list(flatten(lst1)))
print(list(flatten(lst2)))

Output:

[1, 3, 1, 6, 3, 8, 5, 9, 2]
['3', b'A', 0, 1, 4, 0, 1, 2]

I would suggest using generators with yield statement and yield from. Here's an example:

from collections.abc import Iterable

def flatten(items, ignore_types=(bytes, str)):
    """
       Flatten all of the nested lists to the one. Ignoring flatting of iterable types str and bytes by default.
    """
    for x in items:
        if isinstance(x, Iterable) and not isinstance(x, ignore_types):
            yield from flatten(x)
        else:
            yield x

values = [7, [4, 3, 5, [7, 3], (3, 4), ('A', {'B', 'C'})]]

for v in flatten(values):
    print(v)

Considering the list has just integers:

import re
l = [[1, 2, 3], [4, 5, 6], [7], [8, 9]]
list(map(int,re.sub('(\[|\])','',str(l)).split(',')))

def flatten_array(arr):
  result = []
  for item in arr:
    if isinstance(item, list):
      for num in item:
        result.append(num)
    else:
      result.append(item)
  return result

print(flatten_array([1, 2, [3, 4, 5], 6, [7, 8], 9]))
// output: [1, 2, 3, 4, 5, 6, 7, 8, 9]

Simplest Way to do in python without any library

This function will work for even multidimensional list also

using recursion we can achieve any combination of list inside list, we can flatten it without using any library.

#Devil
x = [[1, 2, 3], [4, 5, 6], [7], [8, 9]]


output = []
def flatten(v):
    if isinstance(v, int):
        output.append(v)
    if isinstance(v, list):
        for i in range(0, len(v)):
            flatten(v[i])

flatten(x)
print("Output:", output)
#Output: [1, 2, 3, 4, 5, 6, 7, 8, 9]

#Adding more dimensions 
x = [ [1, [2, 3, [4, 5], [6]], 7 ], [8, [9, [10]]] ]
flatten(x)
print("Output:", output)
#Output: [1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

I created a little function which can basically flatten anything. You can get it with pip: pip install flatten-everything

from flatten_everything import flatten_everything
withoutprotection=list(
    flatten_everything(
        [
            1,
            1,
            2,
            [3, 4, 5, [6, 3, [2, 5, ["sfs", "sdfsfdsf",]]]],
            1,
            3,
            34,
            [
                55,
                {"brand": "Ford", "model": "Mustang", "year": 1964, "yearxx": 2020},
                pd.DataFrame({"col1": [1, 2], "col2": [3, 4]}),
                {"col1": [1, 2], "col2": [3, 4]},
                55,
                {"k32", 34},
                np.array([[[1, 2], [3, 4]], [[5, 6], [7, 8]]]),
                (np.arange(22), np.eye(2, 2), 33),
            ],
        ]
    )
)
print(withoutprotection)
output:
[1, 1, 2, 3, 4, 5, 6, 3, 2, 5, 'sfs', 'sdfsfdsf', 1, 3, 34, 55, 'Ford', 'Mustang', 1964, 2020, 1, 2, 3, 4, 1, 2, 3, 4, 55, 34, 'k32', 1, 2, 3, 4, 5, 6, 7, 8, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 1.0, 0.0, 0.0, 1.0, 33]

You can even protect objects from getting flattened:

from flatten_everything import ProtectedDict,ProtectedList,ProtectedTuple
withprotection=list(
    flatten_everything(
        [
            1,
            1,
            2,
            [3, 4, 5, [6, 3, [2, 5, ProtectedList(["sfs", "sdfsfdsf",])]]],
            1,
            3,
            34,
            [
                55,
                ProtectedDict({"brand": "Ford", "model": "Mustang", "year": 1964, "yearxx": 2020}),
                pd.DataFrame({"col1": [1, 2], "col2": [3, 4]}),
                {"col1": [1, 2], "col2": [3, 4]},
                55,
                {"k32", 34},
                np.array([[[1, 2], [3, 4]], [[5, 6], [7, 8]]]),
                ProtectedTuple((np.arange(22), np.eye(2, 2), 33)),
            ],
        ]
    )
)
print(withprotection)
output:
[1, 1, 2, 3, 4, 5, 6, 3, 2, 5, ['sfs', 'sdfsfdsf'], 1, 3, 34, 55, {'brand': 'Ford', 'model': 'Mustang', 'year': 1964, 'yearxx': 2020}, 1, 2, 3, 4, 1, 2, 3, 4, 55, 34, 'k32', 1, 2, 3, 4, 5, 6, 7, 8, (array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,17, 18, 19, 20, 21]), array([[1., 0.], [0., 1.]]), 33)]

np.hstack(listoflist).tolist()