Python中是否有SciPy函数或NumPy函数或模块来计算给定特定窗口的1D数组的运行平均值?


当前回答

出于教学目的,让我再添加两个Numpy解决方案(比cumsum解决方案慢):

import numpy as np
from numpy.lib.stride_tricks import as_strided

def ra_strides(arr, window):
    ''' Running average using as_strided'''
    n = arr.shape[0] - window + 1
    arr_strided = as_strided(arr, shape=[n, window], strides=2*arr.strides)
    return arr_strided.mean(axis=1)

def ra_add(arr, window):
    ''' Running average using add.reduceat'''
    n = arr.shape[0] - window + 1
    indices = np.array([0, window]*n) + np.repeat(np.arange(n), 2)
    arr = np.append(arr, 0)
    return np.add.reduceat(arr, indices )[::2]/window

使用的函数:as_strided, add.reduceat

其他回答

你可以用以下方法计算运行平均值:

import numpy as np

def runningMean(x, N):
    y = np.zeros((len(x),))
    for ctr in range(len(x)):
         y[ctr] = np.sum(x[ctr:(ctr+N)])
    return y/N

但是速度很慢。

幸运的是,numpy包含一个卷积函数,我们可以用它来加快速度。运行均值相当于将x与一个长度为N的向量进行卷积,其中所有元素都等于1/N。卷积的numpy实现包括起始瞬态,所以你必须删除前N-1点:

def runningMeanFast(x, N):
    return np.convolve(x, np.ones((N,))/N)[(N-1):]

在我的机器上,快速版本要快20-30倍,这取决于输入向量的长度和平均窗口的大小。

请注意,卷积确实包括一个“相同”模式,它似乎应该解决开始的瞬态问题,但它在开始和结束之间分割。

有关现成的解决方案,请参见https://scipy-cookbook.readthedocs.io/items/SignalSmooth.html。 它提供了平窗类型的运行平均值。请注意,这比简单的do-it-yourself卷积方法要复杂一些,因为它试图通过反射数据来处理数据开头和结尾的问题(在您的情况下可能有效,也可能无效……)。

首先,你可以试着:

a = np.random.random(100)
plt.plot(a)
b = smooth(a, window='flat')
plt.plot(b)

使用@Aikude的变量,我编写了一行程序。

import numpy as np

mylist = [1, 2, 3, 4, 5, 6, 7]
N = 3

mean = [np.mean(mylist[x:x+N]) for x in range(len(mylist)-N+1)]
print(mean)

>>> [2.0, 3.0, 4.0, 5.0, 6.0]

高效的解决方案

卷积比直接的方法好得多,但(我猜)它使用FFT,因此相当慢。但是,下面的方法特别适用于计算运行平均值

def running_mean(x, N):
    cumsum = numpy.cumsum(numpy.insert(x, 0, 0)) 
    return (cumsum[N:] - cumsum[:-N]) / float(N)

要检查的代码

In[3]: x = numpy.random.random(100000)
In[4]: N = 1000
In[5]: %timeit result1 = numpy.convolve(x, numpy.ones((N,))/N, mode='valid')
10 loops, best of 3: 41.4 ms per loop
In[6]: %timeit result2 = running_mean(x, N)
1000 loops, best of 3: 1.04 ms per loop

注意numpy。allclose(result1, result2)为True,两个方法等价。 N越大,时间差异越大。

警告:虽然cumsum更快,但会增加浮点错误,这可能导致您的结果无效/不正确/不可接受

这里的评论指出了这个浮点错误问题,但我在回答中让它更明显。

# demonstrate loss of precision with only 100,000 points
np.random.seed(42)
x = np.random.randn(100000)+1e6
y1 = running_mean_convolve(x, 10)
y2 = running_mean_cumsum(x, 10)
assert np.allclose(y1, y2, rtol=1e-12, atol=0)

the more points you accumulate over the greater the floating point error (so 1e5 points is noticable, 1e6 points is more significant, more than 1e6 and you may want to resetting the accumulators) you can cheat by using np.longdouble but your floating point error still will get significant for relatively large number of points (around >1e5 but depends on your data) you can plot the error and see it increasing relatively fast the convolve solution is slower but does not have this floating point loss of precision the uniform_filter1d solution is faster than this cumsum solution AND does not have this floating point loss of precision

我还没有检查这有多快,但你可以试试:

from collections import deque

cache = deque() # keep track of seen values
n = 10          # window size
A = xrange(100) # some dummy iterable
cum_sum = 0     # initialize cumulative sum

for t, val in enumerate(A, 1):
    cache.append(val)
    cum_sum += val
    if t < n:
        avg = cum_sum / float(t)
    else:                           # if window is saturated,
        cum_sum -= cache.popleft()  # subtract oldest value
        avg = cum_sum / float(n)