Freelance data scientist · All case studies

House prices, fitted with gradient descent I wrote.

Linear and multivariate regression for real-estate prices — mean squared error, standardised matrices, and learning weights checked across separate trials. No black-box .fit() as the proof.

Client: Pricing model study. Built by Dilshad Raza.

A pricing model that cannot show the update rule is a demo. The work had to implement linear and multivariate regression with an optimiser I could audit: loss, gradients, scale, and whether the weights were stable across trials.

I coded gradient descent for linear and multivariate house-price regression, standardised the input matrices, tracked MSE, and validated the learned weights on separate trials so a shift in initialisation could not hide.

Hire Dilshad Raza for similar freelance data science, machine learning, and AI automation in the UK, United States, and Australia.