Linear Regression

For any Supervised Machine Learning there are two sets, one is the Training Data Set and the other is Testing Data Set

the attributes in the training data set are called as features $x$ and passed as input to the model, and the model predicts a value $\hat{y}$ which should be close to the training estimate $y$

Learning Function

+−⟲

Link to original

The function $f$ can also be represented as $f_{w,b}(X)=w.x+b$, here $w$ and $b$ are numbers and are used to predict the number $\hat{y}$ from the input features $x$, $w$ is called the parameter weight and $b$ is called bias.

LinearRegressionVisual

+−⟲

Link to original

The above is an example for a Linear Regression model, here the blue line represents a linear function that represents our model prediction.

A linear regression where we have a single variable input\feature is also called *Univariate Linear Regression