When I don't enclose it in a function, the code works, but when I write it as a function, I get an undefined error with x and y values. Could you have an idea ? i've tried many things but I couldn't fix it.

import pandas as pd
import numpy as np
from math import sqrt
import matplotlib.pyplot as plt
from sklearn.preprocessing import MinMaxScaler
from sklearn.preprocessing import LabelEncoder

A=pd.read_excel('example.xlsx')
X=A.iloc[:,0:6]
Y=A.iloc[:,6:7]
X1=A.iloc[:,0:1]
X2=A.iloc[:,1:2]
X3=A.iloc[:,3:6] 

def one_hot_encoding():
            
    data=['apple','banana','orange']
   
    label = LabelEncoder()
    int_data = label.fit_transform(data)
    int_data = int_data.reshape(len(int_data), 1)
    
    onehot_data = OneHotEncoder(sparse=False)
    onehot_data = onehot_data.fit_transform(int_data)
    print("Categorical data encoded into integer values....\n")
    print(onehot_data)
one_hot_encoding()

def normalize_data(x,y):
              
    scaler = MinMaxScaler()
    x=pd.DataFrame(scaler.fit_transform(X),columns=X.columns, index=X.index) 
    y=pd.DataFrame(scaler.fit_transform(Y),columns=Y.columns, index=Y.index) 
      
    return x,y
normalize_data(x,y)

def split_data():

    normalize_data()
    x_train, x_test, y_train, y_test = train_test_split(x,y,train_size=0.85)
    retunr x_train.shape, x_test.shape, y_train.shape, y_test.shape 
             
split_data()

CodePudding user response：

Would’ve helped to see more deets about the error. There is a syntax error somewhere. Make sure you have your references right. E.g. lower case x is different from your upper case X. If you’re calling the normalise_data function make sure you give it the arguments as per the definition. And if you’re taking x, y as arguments to a function it might throw an error when you try to redefine x and y inside the function. Also keep scope in mind. That may also throw an undefined error.

CodePudding user response：

1.) You don't need to Label Encode the variable before One hot encoding. You can directly one hot encode them.

2.) The reason you are getting the error x is not defined is because you are returning x and y from the second function and directly using them in your third function. You have to store them in a variable first and then you can use them.

3.) There is a typo in the third function. It should be return instead of retunr!

Below I have corrected the mistakes and it should work now! Hope it helps!

def one_hot_encoding():
            
    data=['apple','banana','orange']
   
    onehot_data = OneHotEncoder(sparse=False)
    onehot_data = onehot_data.fit_transform(data)
    print("Categorical data encoded into integer values....\n")
    print(onehot_data)
one_hot_encoding()

def normalize_data(x,y):
              
    scaler = MinMaxScaler()
    x=pd.DataFrame(scaler.fit_transform(X),columns=X.columns, index=X.index) 
    y=pd.DataFrame(scaler.fit_transform(Y),columns=Y.columns, index=Y.index) 
      
    return x,y
x1, y1 = normalize_data(x,y)

def split_data():

    normalize_data()
    x_train, x_test, y_train, y_test = train_test_split(x1, y1, train_size=0.85)
    
    print(x_train.shape, x_test.shape, y_train.shape, y_test.shape) 
             
split_data()

Cheers!