This repository contains methods to optimize analytic functions.
oracle.py contains oracle for functions, there deriviates, hessian, etc.
optimization.py contains methods, which can find minimum of functions
To get more information about this methods in russian, open Task.pdf
To run experiments, which depend on datasets, you need to create directory datasets near experiments.py and put there datasets from LIBSVM such as w8a, gisetta and others (all needed datasets you can find in the begining of appropriate experiments).