Skip to content

Latest commit

 

History

History
18 lines (14 loc) · 883 Bytes

README.md

File metadata and controls

18 lines (14 loc) · 883 Bytes
Raw

Stochastic Gradient Descent for SoftSVM

The stochastic gradient descent is used for classification in machine learning in the SoftSVM algorithm.

Features

  • Stochastic Gradient Descent for SoftSVM on MATLAB
  • Hinge loss analysis
  • Binary loss analysis
  • Regularization analysis

Dataset

An 4-dimensional dataset with 1372 data points is used for classification. This dataset is the banknote authentication dataset from the UCI repository.

Documentation

This implementation is part of an article published on Towards Data Science. I encourage you to go over this reading.