sklearn.preprocessing.LabelEncoder 使用

sklearn.preprocessing.LabelEncoder




作用:

將字串陣列進行轉換成固定的數字
以方便之後進行運算

概念類似此

["a" , "b" ,"c"] => [0 ,1 ,2]

code:

import pandas as pd
import numpy as np


from sklearn.preprocessing import LabelEncoder


stringdata = ['a' , 'b' ,'c' , 'd' , 'e' , 'b']

le = LabelEncoder()
le.fit(stringdata)

print("LabelEncoder classes_: \n" , le.classes_)

print("LabelEncoder: \n" , le.transform(stringdata))

print("LabelEncoder inverse_transform \n" , le.inverse_transform([0,1,2,3]))

output:

 LabelEncoder classes_:
 ['a' 'b' 'c' 'd' 'e']
LabelEncoder:
 [0 1 2 3 4 1]
LabelEncoder inverse_transform
 ['a' 'b' 'c' 'd']

運算過程會先將字串進行排序
再依序從0開始匹配一個數字

source code