相当于Python中F#的Seq.scan()方法？

Question

Seq.scan()在Python中是否有类似F#的函数？

我想在没有循环的情况下做一些cumsum()或者cumproduct()某种事情.

Answer 1

不.

def scan(op, seq):
  it = iter(seq)
  result = next(it)
  for val in it:
    result = op(result, val)
    yield result

Answer 2

我认为Ignacio的解决方案几乎是正确的,但需要类型的运算符('a - >'a - >'a)并且不会产生第一个元素.

def scan(f, state, it):
  for x in it:
    state = f(state, x)
    yield state
# test
>>> snoc = lambda xs,x: xs+[x]
>>> list(scan(snoc, [], 'abcd'))
[['a'], ['a', 'b'], ['a', 'b', 'c'], ['a', 'b', 'c', 'd']]
>>> list(scan(operator.add, 0, [1,2,3]))
[1,3,6]

具体来说,类型Seq.scan是

('State -> 'T -> 'State) -> 'State -> seq<'T> -> seq<'State>

Python中的默认方法是scan使用该类型编写

('State -> 'State -> 'State) -> seq<'State> -> seq<'State>

这来自Python指定的方式,reduce默认情况下具有相同的类型.