U 3‰d1ã@shdZddlZddlZddlmZmZddlmZGdd„de ƒZ d d „Z d d „Z ddd„Z ddd„ZdS)a’ Our own implementation of the Newton algorithm Unlike the scipy.optimize version, this version of the Newton conjugate gradient solver uses only one function call to retrieve the func value, the gradient value and a callable for the Hessian matvec product. If the function call is very expensive (e.g. for logistic regression with large design matrix), this approach gives very significant speedups. éNé)Úline_search_wolfe1Úline_search_wolfe2é)ÚConvergenceWarningc@s eZdZdS)Ú_LineSearchErrorN)Ú__name__Ú __module__Ú __qualname__©r r ú:/tmp/pip-unpacked-wheel-zrfo1fqw/sklearn/utils/optimize.pyrsrc KsRt|||||||f|Ž}|ddkr)r r%r&r'TT)NN)Ú__doc__Znumpyrr-ÚfixesrrÚ exceptionsrÚ RuntimeErrorrrr$r3r>r r r r Ús @ö i