如何更改直方图标签图标的颜色？

Question

这是代码。但是，直方图有一个黑色图标作为标签。如何将其更改为红色或“热”颜色的渐变。我知道我将条形颜色更改为渐变“热”的颜色，但是我不想要直方图的黑色标签，因为我想将其与分析图区分开来。

import math, matplotlib.pyplot as plt, random 

def probability(x):

    #wavefunction n=0 evaluated at position x
    psi_0_x=math.exp(-x ** 2 / 2.0) / math.pi ** 0.25

    #probability n=0 to be at position x
    psi_0_x_squared= psi_0_x**2

    return psi_0_x_squared

data_x=[0]
x = 0.0        #starts at position 0
delta = 0.5    #stepsize
trial_steps=1000000

for t in range(trial_steps):

    #displace x by delta
    x_new = x + random.uniform(-delta, delta) 

    #selecciono un numero entre 0 y 1 (incluye acceptance y rejection probability). Metropolis!
    #probabilidad de estar en nuevo sitio/probabilidad de quedarme en el sitio anterior 
    if random.uniform(0.0, 1.0) < probability(x_new)/probability(x):

        #me muevo si la condicion es cierta (está en el accepted range)
        x = x_new 
    data_x.append(x)

#histogram
cm = plt.cm.get_cmap('hot') 
n, bins, patches= plt.hist(data_x, bins=100, normed=True, color='r',label='Histogram')
for height, p in zip(n, patches):
    plt.setp(p, 'facecolor', cm(height))

plt.xlabel('x')
plt.ylabel('$Probability =|\psi_0(x)|^2$')

#general analytical formula
x_grid = [a / 100.0 for a in range(-300,301)]
Prob = [probability(position) for position in x_grid]
plt.plot(x_grid, Prob, linewidth=1.5, color='k', label='Analytical')

plt.title("Position's probability density $|\psi_0(x)|^2$ for a harmonic oscillator.")
plt.savefig('ground_probability_x.png')
plt.legend()
plt.show()

Answer 1

一种解决方案是将生成legend 放在更新hist 的facecolor 之前。另外，我注意到一个小问题，为什么在生成legend 之前保存图形？这是固定的代码和结果：

import math, matplotlib.pyplot as plt, random 

def probability(x):

    #wavefunction n=0 evaluated at position x
    psi_0_x=math.exp(-x ** 2 / 2.0) / math.pi ** 0.25

    #probability n=0 to be at position x
    psi_0_x_squared= psi_0_x**2

    return psi_0_x_squared

data_x=[0]
x = 0.0        #starts at position 0
delta = 0.5    #stepsize
trial_steps=1000000

for t in range(trial_steps):

    #displace x by delta
    x_new = x + random.uniform(-delta, delta) 

    #selecciono un numero entre 0 y 1 (incluye acceptance y rejection probability). Metropolis!
    #probabilidad de estar en nuevo sitio/probabilidad de quedarme en el sitio anterior 
    if random.uniform(0.0, 1.0) < probability(x_new)/probability(x):

        #me muevo si la condicion es cierta (está en el accepted range)
        x = x_new 
    data_x.append(x)

#histogram
cm = plt.cm.get_cmap('hot') 
n, bins, patches= plt.hist(data_x, bins=100, normed=True, color='r',label='Histogram')


plt.xlabel('x')
plt.ylabel('$Probability =|\psi_0(x)|^2$')

#general analytical formula
x_grid = [a / 100.0 for a in range(-300,301)]
Prob = [probability(position) for position in x_grid]
plt.plot(x_grid, Prob, linewidth=1.5, color='k', label='Analytical')
plt.title("Position's probability density $|\psi_0(x)|^2$ for a harmonic oscillator.")
# first legend, then updating the facecolor
plt.legend()
for height, p in zip(n, patches):
    plt.setp(p, 'facecolor', cm(height))
# finally save the figure
plt.savefig('ground_probability_x.png')
plt.show()

希望对您有所帮助。谢谢。