python - 从嵌套列表中删除相似元素

Question

我使用以下代码创建嵌套列表:

from itertools import combinations


def get_nested_list(arr, n, k):
    iterator = 0
    nest = []

    for i in range(1, n):

        for combo in combinations(arr, i):
            odd_num = 0

            for item in combo:
                if item % 2 != 0:
                    odd_num += 1

            if odd_num <= k:
                iterator += 1
                nest.append(combo)

    nest = [list(i) for i in nest]
    return nest


a = [1, 2, 3, 4]
b = len(a)
c = 1

print(get_nested_list(a, b, c))

这是我的输出:

[[1], [2], [3], [4], [1, 2], [1, 4], [2, 3], [2, 4], [3, 4], [1, 2, 4], [2, 3, 4]]

但是，我得到了像 [1,2] 和 [1,4] 这样的元素，它们具有相同的 x 值，[2, 4] 和 [3, 4] 具有相同的 y 值。如何创建具有完全不同的子列表的列表？这就是我想要实现的目标:

[[1], [2], [3], [4], [1, 2], [2, 3], [3, 4], [2, 3, 4]]

Answer 1

与其删除相似的，不如从一开始就不要生成它们。实际上，您不需要组合，您需要重叠的 block (包含多个奇怪项目的 block 除外，您的代码已经考虑到了这一点，我在这里进行了调整)。

def overlapping_chunks(size, sequence):
    """Yield overlapping chunks of sequence."""
    for i in range(len(sequence)-size+1):
        yield sequence[i:i+size]

def powerchunk(sequence):
    """Yield all overlapping chunks of sequence, similar to a powerset."""
    for size in range(1, len(sequence)+1):
        yield from overlapping_chunks(size, sequence)

a = [1, 2, 3, 4]
max_odds = 1
result = [
    chunk for chunk in powerchunk(a)
    if sum(item % 2 != 0 for item in chunk) <= max_odds
    ]
print(result)  # -> [[1], [2], [3], [4], [1, 2], [2, 3], [3, 4], [2, 3, 4]]

相关:Splitting a Python list into a list of overlapping chunks