Simple Linear Regression explained so simply that even a 5yo can understand

Linear regression is an algorithm used to predict or visualise a relationship between two different features/variables. In linear regression tasks, there are two kinds of variables being examined: the dependent variable and the independent variable.

Let us build our first Simple Linear Regression Model and learn along the way by building.

This particular model is called as simple because it has only one independent variable.

Let's start by importing the modules

import matplotlib.pyplot as plt
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression

matplotlib: used to plot the data in a graphical manner
pandas: used for working with the dataset
sklearn: used to split the dataset and then apply the linear regression class onto the data.

Importing the dataset

dataset = pd.read_csv("Salary_Data.csv")
X = dataset.iloc[:, :-1].values
y = dataset.iloc[:, -1].values
# X is taking all the values except the last 
# column whereas y is taking the last value

Here we are using the data containing people's salary and working experience to predict someone's salary based on their experience.

This is what the dataset looks like

Splitting the dataset into training and test set

We need to split the dataset into two models i.e. the test set and the training set.

X_train, X_test, y_train, y_test = train_test_split(
        X,y, test_size = 0.2, random_state = 0)
# X_train contains the independent varibale
# y_train contains the dependent variable

Here we have used the train_test_split function that we imported from sklearn.model_selection.
x and y are the variables, test_size tells the function about the size of the test set.

So, if there exists 100 lines of data, It will be split into following segments,

Training: 80 lines
Testing: 20 lines

Training the model

After we are done with splitting the model, now is the time to actually train the model with the training set.

regressor = LinearRegression()
regressor.fit(X_train, y_train)

We simply initialise the LinearRegression class and then pass our training sets into the fit() method of the LinearRegression Class

Visualising the training set results

plt.scatter(X_train, y_train, color= "red")

# Plotting the data
plt.plot(X_train, regressor.predict(X_train), color="blue" )

# Add title to the plot
plt.title("Salary vs Experience(train)")

# Labels on x and y axis
plt.xlabel("Years of Experience")
plt.ylabel("Salary")

#Finally, display the plot
plt.show()

The output of the following code block will be

Visualising the test set results

plt.scatter(X_test, y_test, color= "red")

# Here we are not replacing X_train with X_test because this line tells us about the data predicted and how close our results are to the training set
plt.plot(X_train, regressor.predict(X_train), color="blue" )

# Add title and labels
plt.title("Salary vs Experience (test)")

plt.xlabel("Years of Experience")
plt.ylabel("Salary")

# Finally, display the plot
plt.show()