CODEWITHSUNDEEP
pythonpandasnumpyscikit-learnsklearn-pandas
NgBrandon
NgBrandon

Reputation: 193

Reverse Label Encoding giving error

I label encoded my categorical data into numerical data using label encoder

data['Resi'] = LabelEncoder().fit_transform(data['Resi'])

But I when I try to find how they are mapped internally using

list(LabelEncoder.inverse_transform(data['Resi']))

I am getting below error

TypeError                                 Traceback (most recent call last)
<ipython-input-67-419ab6db89e2> in <module>()
----> 1 list(LabelEncoder.inverse_transform(data['Resi']))

TypeError: inverse_transform() missing 1 required positional argument: 'y'

How to fix this

Sample data

Resi
IP
IP
IP
IP
IP
IE
IP
IP
IP
IP
IP
IPD
IE
IE
IP
IE
IP
IP
IP

Upvotes: 6

Views: 9821

Answers (3)

Sadabrata Konar
Sadabrata Konar

Reputation: 1

encoder = LabelEncoder()  
encoder.inverse_transform(data['Resi'])

Upvotes: 0

Vivek Kumar
Vivek Kumar

Reputation: 36619

You are not storing the LabelEncoder() object anywhere. You need to save it like this:

le = LabelEncoder()

And then call fit(), or transform().

import pandas as pd
import numpy as np
from sklearn.preprocessing import LabelEncoder

ls = ['IP', 'IP', 'IP', 'IP', 'IP', 'IE', 'IP', 'IP', 'IP', 'IP', 'IP', 'IPD', 'IE', 'IE', 'IP', 'IE', 'IP', 'IP', 'IP']

data = pd.DataFrame(np.array(ls).reshape(-1,1), columns=['Resi'])

le = LabelEncoder()
data['Resi'] = le.fit_transform(data['Resi'])

df['resi'] = LabelEncoder().fit_transform(df['resi'])
list(le.inverse_transform(data['Resi']))

Out: 
['IP',
 'IP',
 'IP',
 'IP',
 'IP',
 'IE',
 'IP',
 'IP',
 'IP',
 'IP',
 'IP',
 'IPD',
 'IE',
 'IE',
 'IP',
 'IE',
 'IP',
 'IP',
 'IP']

Upvotes: 1

jezrael
jezrael

Reputation: 863741

You can check label encoding:

>>> from sklearn import preprocessing
>>> le = preprocessing.LabelEncoder()
>>> le.fit([1, 2, 2, 6])
LabelEncoder()
>>> le.classes_
array([1, 2, 6])
>>> le.transform([1, 1, 2, 6])
array([0, 0, 1, 2])
>>> le.inverse_transform([0, 0, 1, 2])
array([1, 1, 2, 6])

And for your solution:

from sklearn.preprocessing import LabelEncoder

le = LabelEncoder().fit(data['Resi'])
data['Resi'] = le.transform(data['Resi'])
print (data.tail())
    Resi
14     1
15     0
16     1
17     1
18     1

L = list(le.inverse_transform(data['Resi']))
print (L)
['IP', 'IP', 'IP', 'IP', 'IP', 'IE', 'IP', 'IP', 'IP', 
 'IP', 'IP', 'IPD', 'IE', 'IE', 'IP', 'IE', 'IP', 'IP', 'IP']

EDIT:

d = dict(zip(le.classes_, le.transform(le.classes_)))
print (d)
{'IE': 0, 'IPD': 2, 'IP': 1}

Upvotes: 8

Related Questions