-
Notifications
You must be signed in to change notification settings - Fork 0
/
preprocessing.py
51 lines (38 loc) · 1.44 KB
/
preprocessing.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
# -*- coding: utf-8 -*-
"""
Created on Fri Nov 13 14:03:07 2015
@author: abzooba
"""
import numpy as np
from sklearn.preprocessing import LabelEncoder
class FeatureMixer(object):
def __init__(self, dataRover):
self.directory = dataRover.directory
self.train = dataRover.train
self.test = dataRover.test
self.categoricals = dataRover.categoricals
self.numericals = dataRover.numericals
self.targets = dataRover.targets
self.id_col = dataRover.id_col
def outlierCleaning(self):
pass
def categoryEncoder(self):
pass
def missingHandler(self):
self.train = self.train.fillna(0)
self.test = self.train.fillna(0)
def preProcessing(self):
print 'Generic feature mixer...'
# pre-process categoricals
for cate in self.categoricals:
column = self.train[cate]
dtype = str(column.dtype)
if dtype == 'object' and cate not in self.targets:
le = LabelEncoder()
le.fit(np.append(self.train[cate], self.test[cate]))
self.train[cate] = le.transform(self.train[cate])
self.test[cate] = le.transform(self.test[cate])
self.features = self.categoricals + self.numericals
for t in self.targets:
self.features.remove(t)
return True