In [1]:

    

import pandas as pd
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt
sns.set(style="white")
sns.set_context("talk")


    

In [2]:

    

# small example
data = [7, 3, 15]
widths = [1, 5, 3]
left = [0, 1, 6]
plt.bar(left, data, width = widths, color=('orange','green','blue'), 
    alpha = 0.6, align='edge', edgecolor = 'k', linewidth = 2)
plt.show()


    

In [3]:

    

df = pd.read_csv('raw/2016-17-ClassCentral-Survey-data-noUserText.csv', decimal=',', encoding = "ISO-8859-1")


    

In [4]:

    

age = df['What is your age range?'].value_counts()
age


    

    
Out[4]:





46-55 years old    475
36-45 years old    420
56-65 years old    402
66+ years old      285
30-35 years old    253
22-25 years old    203
26-29 years old    199
18-21 years old    176
10-17 years old     51
Name: What is your age range?, dtype: int64

In [5]:

    

from_age =age.index.map(lambda x: x[:2]).astype(np.int)


    

In [6]:

    

to_age = age.index.map(lambda x: x[3:5])
print(to_age == ' y')
to_age.values[to_age == ' y'] = '75' # dummy max age (unix)
# to_age[to_age == ' y'] = '75' # dummy max age (windows)
to_age = to_age.astype(np.int)


    

    




[False False False  True False False False False False]

In [7]:

    

widths = to_age - from_age + 1
widths


    

    
Out[7]:





Int64Index([10, 10, 10, 10, 6, 4, 4, 4, 8], dtype='int64')

In [8]:

    

age_df = pd.DataFrame({'from': from_age, 'to': to_age, 'span': widths, 'count': age})


    

In [9]:

    

sorted_age = age_df.sort_values('from')


    

In [10]:

    

plt.bar(sorted_age['from'], sorted_age['count']/sorted_age['span'], width = sorted_age['span'], 
    alpha = 0.3, align='edge')
plt.xlabel('age of respondents')
plt.show()


    

In [11]:

    

fig = plt.figure(figsize=(10,10))
st = fig.suptitle("Age of respondents")
n = 0
for region in df['Which region of the world are you in?'].dropna().unique():
    dfc = df[df['Which region of the world are you in?'] == region]
    age = dfc['What is your age range?'].value_counts()
    from_age = age.index.map(lambda x: x[:2]).astype(np.int)
    to_age = age.index.map(lambda x: x[3:5])
    to_age.values[to_age == ' y'] = '75' # dummy max age (unix)
    #to_age[to_age == ' y'] = '75' # dummy max age (windows)
    to_age = to_age.astype(np.int)
    widths = to_age - from_age + 1
    age_df = pd.DataFrame({'from': from_age, 'to': to_age, 'span': widths, 'count': age})
    sorted_age = age_df.sort_values('from')
    n += 1
    plt.subplot(4, 3, n)
    plt.bar(sorted_age['from'], sorted_age['count']/sorted_age['span'], width = sorted_age['span'], 
        alpha = 0.3, align='edge')
    plt.yticks([])
    plt.title(region)   
plt.tight_layout(rect=[0, 0, 1, .95])
plt.show()