你的第一段代码以一种高效的（向量）方式完成了你所期望的工作：它将每个值乘以一个介于0.8和1.2之间的不同随机因子。
为了说服自己，只需设计一个小的图形示例：

import pandas as pd
import numpy as np

N = 100

x_test = np.sin(np.linspace(0, 5, N))
noise = np.random.uniform(0.8, 1.2, size=x_test.shape)

x_test_noise = x_test * noise

df = pd.DataFrame({'real': x_test, 'noise': noise, 'real*noise': x_test_noise})
df.plot()

# add lower/upper bounds
df['real'].mul(0.8).plot(ls=':', color='k')
df['real'].mul(1.2).plot(ls=':', color='k')

字符串
输出量：

的数据

绘制没有pandas依赖的代码

import matplotlib.pyplot as plt
import numpy as np

N = 100

x_test = np.sin(np.linspace(0, 5, N))
noise = np.random.uniform(0.8, 1.2, size=x_test.shape)

x_test_noise = x_test * noise

ax = plt.subplot()

ax.plot(x_test, label='real')
ax.plot(noise, label='noise')
ax.plot(x_test_noise, label='real*noise')
ax.plot(x_test*0.8, ls=':', color='k')
ax.plot(x_test*1.2, ls=':', color='k')
plt.legend()

型