Awesome Domain Adaptation Python Toolbox

Regularized Levenberg-Marquardt algorithm for nonlinear regression on small size datasets

This project focuses on curve analysis, including regularizing curves, exploring symmetry, and completing incomplete shapes. It employs techniques like curve fitting, Bézier curve fitting, and symmetry detection to identify regular geometric shapes, detect reflection symmetry in closed curves, and algorithmically fill gaps in occluded shapes.

Multiclass Logistic, Classification Pipeline, Cross Validation, Gradient Descent, Regularization

This program uses the polarity and intensity of the words to assign one of five ratings to product reviews using Multinomial Logistic Regression. The data given was biased with more positive than negative reviews and hence required regularization.