How To Perform Linear Regression With Sklearn?

Linear regression is a statistical approach that models the relationship between input features and output. The input features are called the independent variables, and the output is called a dependent variable. 

In this regression task we will predict the percentage of marks that a student expects to score based upon the number of hours they studied. This is a simple Linear Regression With Sklearn as it involves just two variables.

Simple linear regression mathematically can be given by:

Source: https://medium.com/towards-artificial-intelligence/machine-learning-algorithms-for-beginners-with-python-code-examples-ml-19c6afd60daa

Tool to be use: Jupyter , Excel and python library whenever needed

Linear Regression With Sklearn

Since we will going to use various library so we need to import them, before that lets us know use of these library first

1. Numpy: It is a math library to work with n-dimensional arrays in Python. It enables us to do mathematical computation over given data in very efficient way.
2. Scipy: this library used in many ways. Scipy is a functional library for scientific and high-performance computations.
3. Matplotlib: It is a trendy plotting package that provides plotting of the chart for two or three dimensions.
4. Scikit-learn: Sklearn is most used library in machine learning as it has various function to perform classification, regression and clustering algorithm.

Another necessary step to start with is reading the dataset present in the .xlsx or .csv format.

Now we have to select the feature so in this feature is Scores and Hours as this is very simple dataset.

Before selecting the feature if there is any noise than we have to clean it.

After that will start with Linear Regression With Sklearn.

So in this we have to select x and y value which denote feature and target value

So in general term our target ‘y’ is always stay at he last column so we can apply below method for simplicity

Now we have to split the data into test and train data

Than we have to apply linear regression function into this

After this we have to fit the data by using fit function

so, After this we have to check the coefficient of determination 

After knowing the coefficient of determination lets find out b0 value which is intercept

Now lets check another factor which is important i.e. slope

 then, let’s plot the graph.

Let’s check the algorithm by predicting the some test data value

To check predicted score we can apply below method

 We can compare actual and predicted value for better understanding of the accuracy

Lets check for predicted score of the student who studied for 9.25 hrs/day

Now if we check the graph we can see linear graph with most value line up there

we can check the predicted and observed error using metrics

It may happen that the data that got collected cannot be performed using linear regression. Most of the time data follow polynomial trend where data are more ways in the non linear manner

Source: https://medium.com/towards-artificial-intelligence/machine-learning-algorithms-for-beginners-with-python-code-examples-ml-19c6afd60daa

Written By: Nikesh Maurya

Reviewed By: Krishna Heroor

If you are Interested In Machine Learning You Can Check Machine Learning Internship Program
Also Check Other Technical And Non Technical Internship Programs