Linear Regression
Here’s some info about Linear Regression, It is an approach to form a relation between one or more explanatory variables in a linear fashion. Those variables may be dependent or independent. Multiple Linear Regression is used to make predictions based on multiple variables. It assumes a linear relationship between input variable (x) and the output variable (y), Where this y is formed by the linear combination of the input variables.
The equation for linear regression is y = b0+b1*x , Where b0,b1 are changed to adjust the line to get the best possible prediction in order to minimize loss.
As we know that y=mx+c is the equation of a line with slope “m” and intercept “c”. Here slope is the weights and intercept is the bias. So the actual equation can be written as y=wx+b.
From above figure, you can observe the best possible line to predict the value of y based on X. So, Linear Regression is nothing but finding a best possible line which makes better approximate prediction with minimum loss. The figure shows simple linear regression because it is forming a linear line based on single input variable but if there is more than one variable then it is called multivariate linear regression.
The loss function used here is MSE (mean square Error). It is calculated based on the ground truth and the predicted value. So this is a high level intro about Linear Regression.
Thanks for going through my article !!!
