In [30]:

    

import numpy as np


    

In [48]:

    

%run k_means_cluster.py


    

In [12]:

    

exercised_stock_options = []
for person in data_dict:
    if data_dict[person]['exercised_stock_options'] != 'NaN':
        exercised_stock_options.append(int(data_dict[person]['exercised_stock_options']))
        
print min(exercised_stock_options)
print max(exercised_stock_options)


    

    




3285
34348384

In [13]:

    

salaries = []
for person in data_dict:
    if data_dict[person]['salary'] != 'NaN':
        salaries.append(int(data_dict[person]['salary']))
        
print min(salaries)
print max(salaries)


    

    




477
1111258

In [44]:

    

features_list


    

    
Out[44]:





['poi', 'salary', 'exercised_stock_options']

In [40]:

    

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(finance_features)


    

    
Out[40]:





MinMaxScaler(copy=True, feature_range=(0, 1))

In [45]:

    

def scale(number, min_, max_):
    if min_ != max_:
        return float(number-min_)/(max_-min_)
    else:
        return 1


    

In [46]:

    

scale(200000, 477, 1111258)


    

    
Out[46]:





0.17962406631010072

In [47]:

    

scale(1e6,3285,34348384)


    

    
Out[47]:





0.029020588934683227

In [ ]: