使用元组元素从列表制作频率直方图

Question

我想做一个词频分布，x轴是词数，y轴是频数。

我有以下列表：

example_list = [('dhr', 17838), ('mw', 13675), ('wel', 5499), ('goed', 5080), 
                ('contact', 4506), ('medicatie', 3797), ('uur', 3792),
                ('gaan', 3473), ('kwam', 3463), ('kamer', 3447), 
                ('mee', 3278), ('gesprek', 2978)]

我尝试先将其转换为 pandas DataFrame，然后使用 pd.hist() 如下例所示，但我无法理解并认为它实际上很简单但可能我遗漏了什么。

import numpy as np
import matplotlib.pyplot as plt

word = []
frequency = []

for i in range(len(example_list)):
  word.append(example_list[i][0])
  frequency.append(example_list[i][1])


plt.bar(word, frequency, color='r')
plt.show()

Answer 1

您不能将 word 传递到 matplotlib.pyplot.bar directly. However you could create an indices array for bar and then replace these indices with the words using matplotlib.pyplot.xticks:

import numpy as np
import matplotlib.pyplot as plt

indices = np.arange(len(example_list))
plt.bar(indices, frequency, color='r')
plt.xticks(indices, word, rotation='vertical')
plt.tight_layout()
plt.show()

创建 word 和 frequency 的 for 循环也可以替换为简单的 zip 和列表解包：

word, frequency = zip(*example_list)

Answer 2

使用pandas：

import pandas as pd
import matplotlib.pyplot as plt

example_list = [('dhr', 17838), ('mw', 13675), ('wel', 5499), ('goed', 5080), ('contact', 4506), ('medicatie', 3797), ('uur', 3792), ('gaan', 3473), ('kwam', 3463), ('kamer', 3447), ('mee', 3278), ('gesprek', 2978)] 

df = pd.DataFrame(example_list, columns=['word', 'frequency'])
df.plot(kind='bar', x='word')

使用元组元素从列表制作频率直方图

Make Frequency histogram from list with tuple elements

python

plot

distribution

frequency

matplotlib