#include <toolkits/factorization/loss_model_profiles.hpp>

Detailed Description

The base class for the generative models. These models encapsulate the part of the problem surrounding the translation of the underlying linear model to the target/response variable. Thus it encapsulates the (1) translation function from linear model to response and (2) the loss function used to fit the coefficients of the linear model to predict the response.

To make reporting easier, report_loss translates a cumulative loss value – the sum of loss(...) over all data points – to a standard error measure. It's name is returned along with the reportable value.

Definition at line 32 of file loss_model_profiles.hpp.

The documentation for this class was generated from the following file:

toolkits/factorization/loss_model_profiles.hpp