dl::loss::HingeLoss< T > Class Template Reference

Hinge Loss with autograd support. More...

#include <losses.hpp>

Inheritance diagram for dl::loss::HingeLoss< T >:

Collaboration diagram for dl::loss::HingeLoss< T >:

Public Member Functions
	HingeLoss (const std::string &reduction="mean")
	Constructor.
Variable< T >	forward (const Variable< T > &predictions, const Variable< T > &targets) override
	Forward pass: compute hinge loss.
Public Member Functions inherited from dl::loss::AutogradLoss< T >
virtual	~AutogradLoss ()=default
Variable< T >	operator() (const Variable< T > &predictions, const Variable< T > &targets)
	Convenience operator for computing loss.

Detailed Description

template<typename T>
class dl::loss::HingeLoss< T >

Hinge Loss with autograd support.

Hinge(y_pred, y_true) = max(0, 1 - y_true * y_pred)

Commonly used for SVM and margin-based classification.

Definition at line 176 of file losses.hpp.

Constructor & Destructor Documentation

HingeLoss()

template<typename T >

dl::loss::HingeLoss< T >::HingeLoss ( const std::string &  reduction = "mean" )

inlineexplicit

Constructor.

Parameters

reduction

Type of reduction ('mean', 'sum', 'none')

Definition at line 182 of file losses.hpp.

Member Function Documentation

forward()

template<typename T >

Variable< T > dl::loss::HingeLoss< T >::forward ( const Variable< T > &  predictions, const Variable< T > &  targets  )

overridevirtual

Forward pass: compute hinge loss.

Parameters

predictions	Predicted values
targets	Target values (-1 or +1)

Returns

Hinge loss

Implements dl::loss::AutogradLoss< T >.

Definition at line 137 of file losses.cpp.

---

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

• include/loss/losses.hpp
• src/loss/losses.cpp