Implementation of Support vector regression algorithm in Python (SVR):

Support Vector Regression (SVR) is similar to Support Vector machine (SVM). But SVM is used to predict discrete class labels and SVR is used to predict continuous class labels.

Support Vector Machine

Non-Linear Support Vector Regression

Linear Support Vector Regression

Support Vector regression can be used for both linear and non linear regressions. Epsilon is the distance around the hyperplanes. The points inside the epsilon are considered as correct predictions. So epsilon can be tuned to increase the performance.
Slack is the distance above upper hyperplane and below lower hyperplane, till which the regression errors allow the points to exist. But it satisfies all required conditions.

Implementation of SVR:

Importing libraries:

Loading the data:

First five rows:

Overall information:

Checking for null values:

Statistically describing the data:

Dropping the column Position:

 Independent and Dependent variables:

Standardizing the data:

Fitting the model:

Visualizing SVR: