Join GitHub today
GitHub is home to over 31 million developers working together to host and review code, manage projects, and build software together.Sign up
LMNN: optimization by avoiding impostor recalculation using bounds #1448
This is related to #1407 and is an optimization that we could do. It focuses on the recalculation of impostors that is performed at each iteration (which happens every
Essentially to implement this, we will have to do the following:
Note that the bound in the paper can apply for each point individually, so it is possible to do the following:
I tried these bound by creating an overload of Impostors() to calculate impostors for some specific points of the dataset which violates the bounds (there is no tree caching as of now).