Data Preprocessing in Machine Learning using Python - SimpleImputer, OneHotEncoder, train_test_spli

GitHub: https://github.com/bibhutighimire/Data-Preprocessing-in-Machine-Learning-using-Python

Google Collab: https://colab.research.google.com/drive/1e1sKsVO-26kicMXbdq6aoe_shlgsxWh #scrollTo=fGOiYH95ojot

Data Preprocessing in Machine Learning using Python 2023

#SimpleImputer #OneHotEncoder #train_test_split #StandardScaler #LabelEncoder

Website: https://www.javatpoint.com/how-to-get-datasets-for-machine-learning

//Import Library:
import pandas as pd
import numpy as np

//Import Dataset:
dataset = pd.read_csv('Data.csv')
print(dataset)

//Split dataset into X and y i.e. independent and dependent model
X = dataset.iloc[:,:-1].values
print(X)

y = dataset.iloc[:,-1].values
print(y)

//Handling missing data:
from sklearn.impute import SimpleImputer

si = SimpleImputer(missing_values= np.nan , strategy='mean')
si.fit(X[:,1:3])
X[:, 1:3] = si.transform(X[:,1:3])
print(X)

//Encoding categorical data:
from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import OneHotEncoder

ct = ColumnTransformer(transformers=[('encoder', OneHotEncoder(), [0])] , remainder= 'passthrough')

X = ct.fit_transform(X)
print(X)

from sklearn.preprocessing import LabelEncoder

le = LabelEncoder()
y = le.fit_transform(y)
print(y)

//Splitting data into training and testing model:
from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)

print(X_train)
print(y_train)
print(y_test)
print(X_test)

//Feature Scaling:
from sklearn.preprocessing import StandardScaler

ss = StandardScaler()
X_train[:,3:5] = ss.fit_transform(X_train[:,3:5])
X_test[:,3:5] = ss.transform(X_test[:,3:5])

print(X_train)
print(X_test)