如何从 python 中的数组列表中更快地访问字典的值？

看看你的代码，我认为你应该使用而不是字典。另外，将末尾的数字转换为字符串：my_dictnp.array

# convert my_dict to np.array
arr = np.array([my_dict[i] for i in range(1, max(my_dict) + 1)])

# create the final set:
out = {frozenset(arr[l - 1]) for l in my_list}

# optionally, convert to string:
out = [" ".join(map(str, s)) for s in out]
print(out)

指纹：

['1 2 3', '0 1 3', '0 1 2', '0 1 2 3']

编辑：将您的文件发布在问题中，您可以尝试（在我的机器上，原始答案需要 11 秒和 19 秒）：

import ast

file_list = "list_array.txt"
file_dictionary = "dictionary_example.txt"

with open(file_dictionary, "r") as file_dict:
    my_dict = ast.literal_eval(file_dict.read())

# convert keys back to string (to not convert the lines to int)
my_dict = {str(k): v for k, v in my_dict.items()}

out = set()
with open(file_list, "r") as file:
    for line in file:
        out.add(frozenset(my_dict[i] for i in line.split()))

out = [" ".join(map(str, s)) for s in out]
print(out)

编辑2：两种方法的比较：

import ast
from timeit import timeit

# read sample data:

file_list = "list_array.txt"
file_dictionary = "dictionary_example.txt"

with open(file_dictionary, "r") as file_dict:
    my_dict = ast.literal_eval(file_dict.read())

my_list = []
with open(file_list, "r") as file:
    for line in file:
        my_list.append(np.array(list(map(int, line.split()))))


def fn1(my_list, my_dict):
    arr = np.array([my_dict[i] for i in range(1, max(my_dict) + 1)])
    out = {frozenset(np.unique(arr[l - 1])) for l in my_list}
    return [" ".join(map(str, s)) for s in out]


def fn2(my_list, my_dict):
    out = {frozenset(np.unique(np.vectorize(my_dict.get)(l))) for l in my_list}
    return [" ".join(map(str, s)) for s in out]


assert len(fn1(my_list, my_dict)) == len(fn2(my_list, my_dict))

t1 = timeit("fn1(my_list, my_dict)", number=1, globals=globals())
t2 = timeit("fn2(my_list, my_dict)", number=1, globals=globals())

print(t1)
print(t2)

指纹：

1.7291587160434574
11.316183750052005