Skip to content

hackathonsforschools/linear-regression-workshop

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

3 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

This is a workshop to showcase the way in which a trend line can be calculated to fit a data set. This trend line can then be used to make predictions.

Making predictions from a data model is a crucial part of Artificial Intelligence and Machine Learning.

What is linear regression?

Linear regression is a way to explore the relationship between a dependent variable and an independent one.

Explanation of terms:

Epochs - the number of times the process is run on the data set, more repeats results in a more accurate model. An excessive number of repeats can result in overfitting, which means the model is specific to only the data present, and can not neccesarily make accurate predictions.

Weight - the coefficient of a variable in the resulting equation.

Learning Rate - the rate at which the gradient of the line is altered. Setting this too low results in convergence being too slow. Setting it too high means the changes will be unstable.