Question

这些是我使用的行，它生成一个列表，列出了总和为输入的整数，然后说明了组合的数量。

import itertools

val = int(input())
r = range(val)[1:]
res = []    
for i in range(len(r)+1):
    res += [list(x) for x in itertools.combinations(r, i) if sum(list(x)) == val]    
print("Solution : %s " % res)
print("Combinations : %s" % len(res))

Answer 1

看起来您正在计算每个可能的组合，然后只保留满足您的总和条件的组合。它正在搜索的组合池以2 ^（n-1）的速度增长，您可以通过注释条件部分来查看：

import itertools
import time

startTime = time.time()

val = 21
r = range(val)[1:]
res = []    
for i in range(len(r)+1):
    res += [list(x) for x in itertools.combinations(r, i)]# if sum(list(x)) == val]    
#print("Solution : %s " % res)
print("Combinations : %s" % len(res))

print(time.time() - startTime, 'seconds')

试试这个：

def a(lst, target, with_replacement=False):
    def _a(idx, l, r, t, w):
        if t == sum(l): r.append(l)
        elif t < sum(l): return
        for u in range(idx, len(lst)):
            _a(u if w else (u + 1), l + [lst[u]], r, t, w)
        return r
    return _a(0, [], [], target, with_replacement)

val = 50
s = range(1, val)
solutions = a(s, val)
print(solutions)
print(len(solutions))

由thefourtheye代码修改： Finding Combinations to the provided Sum value

计时：原始代码：

val = 24, t = 6.5917372703552253 seconds
val = 25, t = 13.705767631530762 seconds
val = 26, t = 27.934977531433105 seconds
val = 50, t = 459889248.39282094 seconds

上述方法：

val = 24, t = 0.0084612369537353 seconds
val = 25, t = 0.0069310665130615 seconds
val = 26, t = 0.0085859298706054 seconds
val = 50, t = 0.4947729110717773 seconds

Answer 2

可能有很多方法可以提高速度。我试过（并注意到它有很大的不同）一个简单的方法：摆脱生成器和列表之间的重复转换。这是您的代码，已修改：

import time
import itertools

#val = int(input("val:"))
#r = range(val)[1:]

def simple_time_func(func):
    def wrapper(*args, **kwargs):
        start = time.time()
        r = func(*args, **kwargs)
        print("{} took {:.3f} seconds.".format(func.__name__, time.time() - start))
        return r
    return wrapper

@simple_time_func
def original(val, r):
    res = list()
    for i in range(len(r) + 1):
        res += [list(x) for x in itertools.combinations(r, i) if sum(list(x)) == val]
    return res

@simple_time_func
def improved0(val, r):
    res = list()
    for i in range(len(r) + 1):
        res += [x for x in itertools.combinations(r, i) if sum(x) == val]
    return res


for val in range(21, 28):
    r = range(val)[1:]
    print("\nval: {}".format(val))
    for func in (original, improved0):
        res = func(val, r)
        #print("    Solution : %s " % res)
        #print("    Combinations : %s" % len(res))

备注：

我拿了昂贵的部分（for循环）并将其封装在一个函数（original）中以便计时

我将改进的循环放在另一个函数中（称为improved0）

其余代码（不幸的是，其中大部分代码）仅用于性能测试

<强>输出：

(py35x64_test) c:\Work\Dev\StackOverflow\q46853086>"c:\Work\Dev\VEnvs\py35x64_test\Scripts\python.exe" a.py val: 21 original took 0.573 seconds. improved0 took 0.325 seconds. val: 22 original took 1.185 seconds. improved0 took 0.633 seconds. val: 23 original took 2.388 seconds. improved0 took 1.300 seconds. val: 24 original took 4.880 seconds. improved0 took 2.609 seconds. val: 25 original took 9.796 seconds. improved0 took 5.243 seconds. val: 26 original took 19.701 seconds. improved0 took 10.489 seconds. val: 27 original took 39.314 seconds. improved0 took 21.559 seconds.

正如输出显示的那样，这种明显微不足道的变化几乎将时间缩短了一半下一步包括真实优化（因为计算复杂性仍然看起来是指数级的），关于数学和编程技巧的有效使用，更进一步提高速度（不幸的是我已经没时间了）

我的程序需要很长时间才能打印超过25的结果，有没有办法让我在运行时更有效率？

2 个答案: