python 绘图 RecursionError : maximum recursion depth exceeded in comparison

Question

我尝试运行下面的代码

import math
import random
from matplotlib import pyplot as plt
from collections import Counter

def bucketize(point,bucket_size):
    return bucket_size * math.floor(point/bucket_size)

def make_histogram(points, bucket_size):
    return Counter(bucketize(point,bucket_size) for point in points)

def plot_histogram(points,bucket_size,title=""):    
    histogram = make_histogram(points,bucket_size)
    plt.bar(histogram.keys(),histogram.values(),width=bucket_size)
    plt.title(title)
    plt.show()

def inverse_normalCDF(p,miu=0,sigma=1,tolerance=0.00001):
    if miu !=0 or sigma !=1:
        return miu + sigma * inverse_normalCDF(p,tolerance=tolerance)
    low_z,low_p = -10.0,0
    hi_z,hi_p = 10.0,1
    while hi_z - low_z > tolerance:
        mid_z = (low_z + hi_z)/2
        mid_p = normalCDF(mid_z)
        if mid_p > p:
            low_z,low_p = mid_z,mid_p
        elif mid_p > p:
            hi_z, hi_p = mid_z,mid_p
        else:
            break
    return mid_z

def normalCDF(x,miu=0,sigma=1):
    return (1+math.erf((x-miu)/math.sqrt(2)/sigma))/2


random.seed(0)
#uniform = [200*random.random()-100 for _ in range (10000)]
#plot_histogram(uniform,10,"uniform histogram")
normal = [57 * inverse_normalCDF(random.random() for _ in range(10000))]
plot_histogram(normal,10,"normal histogram")

但程序显示错误“RecursionError:比较时超出最大递归深度”，我应该如何修复 RecursionError，因为它会影响 inverse_normalCDF 函数？

错误不再显示函数中超出的最大递归深度，而是

Traceback (most recent call last):
  File "C:\Users\asus\Documents\Sublime\dataScience\normalHistogram.py", line 41, in <module>
    normal = [57 * inverse_normalCDF(random.random() for _ in range(10000))]
  File "C:\Users\asus\Documents\Sublime\dataScience\normalHistogram.py", line 26, in inverse_normalCDF
    if mid_p > p:
TypeError: '>' not supported between instances of 'float' and 'generator'
[Finished in 0.7s]

我使用的python版本是3.7，我使用的matplotlib版本是3.0.3

*更新，改变了

def inverse_normalCDF(p,miu=0,sigma=1,tolerance=0.00001):fix some typos in the code

Answer 1

你快到了。该错误可以通过重写您的列表理解来解决:

normal = [57 * inverse_normalCDF(random.random()) for _ in range(10000)]

我只是移动了一个右括号，以便函数 inverse_normalCDF 为每次迭代传递一个值，而不是整个生成器。