Skip to content

Latest commit

 

History

History
6 lines (4 loc) · 607 Bytes

README.md

File metadata and controls

6 lines (4 loc) · 607 Bytes

Support_Vector_Machines

SVM algorithm for handwritten digit classification

Solving the SVM optimization problem for handwritten digit classification using a general purpose convex optimization package as well as using a customized solver known as LIBSVM. The general purpose convex optimization package used is CVXOPT and LIBSVM can be found here. Linear and Gaussian Kernel are compared for binary classification and Gaussian Kernel is used for multiclass classification.

The complete problem statement is included (ass2_partb_v1.pdf)