如何使用scipy gaussian_kde获得概率密度函数？

Question

I have a 1D data set which is saved in a 1D list. What is the best way to get the probability density function? I tried the usual approach of using scipy gaussian_kde.

array = np.array(values)
kde = gaussian_kde(array)
x = np.linspace(0, 50, 500)
plt.plot(x, kde(x), label="", color="blue")
plt.legend(loc='best')
plt.show()

Produced graph is not the expected probability density function, since probability density functions should have a value between 0 and 1 for each x.

Thanks

Answer 1

Use the following code.

import os
import matplotlib.pyplot as plt
import sys
import math
import numpy as np
import scipy.stats as st
from scipy.stats._continuous_distns import _distn_names
from scipy.optimize import curve_fit

def get_pdf(latency_list):
    np_array = np.array(latency_list)  # convert the list into a numpy array
    ag = st.gaussian_kde(np_array)  # calculate the kernel density function for the latency values
    # list of equidistant values in the range of the latency values
    x = np.linspace(min(latency_list), max(latency_list), (max(latency_list) - min(latency_list)) * 10)
    y = ag(x)  # evaluate the latency values for each x value
    return x, y