forked from ayushsubedi/kohokoho
-
Notifications
You must be signed in to change notification settings - Fork 1
/
kohokoho.py
209 lines (193 loc) · 6.51 KB
/
kohokoho.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
import click
import string
import random
import time
import pandas as pd
from faker import Faker
fake = Faker('en')
class anon(object):
'''Initialize a df as an anon object.
Args:
df: pandas dataframe
Returns:
anon object
'''
def __init__(self, df):
self.original = df.copy()
self.df = df
def anon_name(self, col):
''' Replace entries in name column with fake names generated
from faker.
Args:
col: column containing name data
Returns:
anon object with replaced name col
'''
unique = self.df[col].unique()
map_dict = {_: fake.name() for _ in unique}
self.df[col] = self.df[col].map(map_dict)
def anon_id(self, col):
''' Replace entries in id column with fake uuid generated
from faker.
Args:
col: column containing id data
Returns:
anon object with replaced id col
'''
unique = self.df[col].unique()
map_dict = {_: fake.uuid4() for _ in unique}
self.df[col] = self.df[col].map(map_dict)
def anon_discrete_num(self, col):
''' Replace entries in column with whole nums to random whole
nums in the same range.
Args:
col: column containing whole number data
Returns:
anon object with replaced discrete col
'''
X_std = (self.df[col] - self.df[col].min()) / (
self.df[col].max() - self.df[col].min())
X_scaled = (X_std * (10 - 1) + 1)
X_scaled_randomized = (X_scaled * random.randint(1, 10)).astype(int)
self.df[col] = X_scaled_randomized
def anon_continuous_num(self, col):
''' Replace entries in columns with continuous nums to random whole
nums in the same range.
Args:
col: column containing continuous number data
Returns:
anon object with replaced continuous col
'''
X_std = (self.df[col] - self.df[col].min()) / (
self.df[col].max() - self.df[col].min())
X_scaled = (X_std * (10 - 1) + 1)
X_scaled_randomized = round(X_scaled * random.randint(1, 10), 3)
self.df[col] = X_scaled_randomized
def anon_category(self, col):
''' Replace entries in column with categorical data to
anonymized category.
Args:
col: column containing categorical data
Returns:
anon object with replaced categorical col
'''
unique = self.df[col].unique()
rand_ = random.randint(0, 1000)
map_dict = {
category: "Category_" + str(rand_) + " " + str(i)
for i, category in enumerate(unique)
}
self.df[col] = self.df[col].map(map_dict)
def anon_date(self, col):
''' Replace entries in date column with random date
in the same range.
Args:
col: column containing date data
Returns:
anon object with replaced date col
'''
self.df[col] = pd.to_datetime(
self.df[col], infer_datetime_format=True)
start_date = self.df[col].min()
end_date = self.df[col].max()
map_list = [fake.date_between(
start_date=start_date,
end_date=end_date) for i in range(self.df.shape[0])]
self.df[col] = map_list
def anon_email(self, col):
''' Replace entries in email column with random emails.
Args:
col: column containing email
Returns:
anon object with replaced email col
'''
unique = self.df[col].unique()
map_dict = {_: (''.join(random.choices(
string.ascii_lowercase + string.digits,
k=12)))+'@anonemail.com' for _ in unique}
self.df[col] = self.df[col].map(map_dict)
def save_anon_csv(self):
'''Save anon object to a csv file'''
self.df.to_csv(str(time.time())+'kohokoho.csv', index=False)
def anon_df(self):
return self.df
def _df(self):
return self.original
@click.command()
@click.option(
'--csv',
prompt='Enter location of CSV',
help='Enter a valid filepath or buffer')
def cli(csv):
df = pd.read_csv(csv)
koho_df = anon(df)
click.echo('Columns info: '+str(df.info()))
# name
name_col = click.prompt(
'Enter column/s which stores names, each column separated by a comma',
type=str,
default='')
if (name_col != ''):
for col in name_col.split(","):
koho_df.anon_name(col.strip())
# id
id_col = click.prompt(
'Enter column/s which stores id, each column separated by a comma',
type=str,
default='')
if (id_col != ''):
for col in id_col.split(","):
koho_df.anon_id(col.strip())
# continuous values
continuous_col = click.prompt(
'''Enter column/s which stores continuous numbers,
each column separated by a comma''',
type=str,
default='')
if (continuous_col != ''):
for col in continuous_col.split(","):
koho_df.anon_continuous_num(col)
# discrete_col
discrete_col = click.prompt(
'''Enter column/s which stores discrete numbers,
each column separated by a comma''',
type=str,
default='')
if (discrete_col != ''):
for col in discrete_col.split(","):
koho_df.anon_discrete_num(col)
# category
category_col = click.prompt(
'''Enter column/s which stores categorical values,
each column separated by a comma''',
type=str,
default='')
if (category_col != ''):
for col in category_col.split(","):
koho_df.anon_category(col)
# date
date_col = click.prompt(
'Enter column which stores dates, each column separated by a comma',
type=str,
default='')
if (date_col != ''):
for col in date_col.split(","):
koho_df.anon_date(date_col)
# email
email_col = click.prompt(
'Enter column which stores email, each column separated by a comma',
type=str,
default='')
if (email_col != ''):
for col in email_col.split(","):
koho_df.anon_email(email_col)
# original dataset
click.echo('Original dataset')
click.echo(koho_df._df().head(10))
# final dataset
click.echo('Kohoko dataset')
click.echo(koho_df.anon_df().head(10))
# save anon dataset
if click.confirm('Do you want to save the anonymized csv?'):
koho_df.save_anon_csv()
click.echo('Done!')