Word counts

In [46]:

    

import matplotlib as mpl
import matplotlib.pyplot as plt

%matplotlib inline
mpl.style.use('bmh')


    

In [2]:

    

import pandas as pd


    

In [22]:

    

df = pd.read_csv('data/token-counts.csv').sort_values('token_count')


    

In [45]:

    

x = df['token_count']
y = df['count']


    

In [44]:

    

plt.figure(figsize=(20, 10))

plt.title('Document lengths')
plt.xlabel('Word count')
plt.ylabel('Number of documents')
plt.yscale('log')
plt.xlim((0, 200000))
plt.scatter(x, y, s=5, c='r')
plt.show()