如何计算 scipy 中散点图的相关系数

Question

我正在尝试用 scipy 计算散点图的相关系数，问题是，我在 ndarray 中有一种复杂的数据集，基本语法对我不起作用...

这是我的完整代码：

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import matplotlib.patches as mpatches
from matplotlib.pyplot import figure

figure(figsize=(12, 10), dpi=200)


import scipy.stats
from scipy.stats import t

plt.close('all')


data = np.array([
    [22.8, 14.4],
    [19.6, 3.6],
    [0.3, 16.6],
    [8.9, 7],
    [13.7, 13.4],
    [14.7, 1.5],
    [1.9, 0.4],
    [-1.8, 0.3],
    [-3, -15.3],
    [-5.9, -6.3],
    [-13.4, -15],
    [-5.7, -34.8],
    [-6.8, -12.9],

]) 

custom_annotations = ["K464E", "K472E", "R470E", "K464A", "M155E", "K472A", "M155A", "Q539A", "M155R", "D244A", "E247A", "E247R", "D244K"]
class_colours = ["r", "r", "r", "r", "r", "r", "g", "g", "b", "b", "b", "b", "b"]

for i, point in enumerate(data): 
    plt.scatter(point[0], point[1], marker='o', label=custom_annotations[i], c=class_colours[i], edgecolors='black', linewidths=1, alpha=0.75)
    plt.annotate(custom_annotations[i], (data[i,0], data[i,1]))

plt.xlabel(r'$\Delta  V_{0.5}$  Apo wild-type mHCN2 (mV)', fontsize=10)
plt.ylabel(r'$\Delta \psi$  cAMP-bound wild-type mHCN2 (mV)', fontsize=10)
plt.title('$\Delta \psi$  cAMP-bound wild-type mHCN2 (HHU) vs Change in relative current (Jena)', fontsize=10)

plt.axvline(0, c=(.5, .5, .5), ls= '--')
plt.axhline(0, c=(.5, .5, .5), ls= '--')

scipy.stats.pearsonr(data[i,0], data[i,1])

plt.legend(ncol=3, loc=(1.04,0))
plt.show()

Answer 1

pearsonr 适用于您的数据

scipy.stats.pearsonr(data[:,0], data[:,1]) #change i to : to get the whole col.
# this returns (r_coeff, p_value)

如错误所述，您传递了两个浮点数（即第 i 行的值），但是 corr 需要两个数组，在您的例子中是两列。

如何计算 scipy 中散点图的相关系数

how to calculate correlation coefficient for a scatter-plot in scipy

python

numpy

matplotlib

scipy