TSF_SupervisedMl_480p.mp4
During my internship at The Sparks Foundation, I was tasked with predicting students' percentages based on the number of hours they studied. This project involved a straightforward dataset with just two variables: hours and scores. Instead of relying on external libraries like scikit-learn, I opted to develop my own linear regression algorithm to tackle the task.
I'm pleased to share that through this approach, I successfully minimized the mean absolute error to 4.97, with an alpha (speed of convergence) of around 1.0e-2.
For a deeper insight into my methodology and implementation, I've provided a link to the GitHub repository.
github_repository: https://github.com/ANSHPG/LinearLuminary
Thank you for your interest!