Deep Learning Algorithm Implementations 1.0.0
C++ implementations of fundamental deep learning algorithms
Loading...
Searching...
No Matches
utils::SubFunction< T > Class Template Reference

Subtraction function. More...

#include <autograd.hpp>

Inheritance diagram for utils::SubFunction< T >:
Collaboration diagram for utils::SubFunction< T >:

Public Member Functions

Matrix< T > forward (const std::vector< Variable< T > > &inputs) override
 Forward pass computation.
 
std::vector< Matrix< T > > backward (const Matrix< T > &grad_output) override
 Backward pass computation.
 
- Public Member Functions inherited from utils::Function< T >
virtual ~Function ()=default
 
virtual void save_for_backward (const std::vector< Matrix< T > > &tensors)
 Set saved tensors for backward pass.
 

Additional Inherited Members

- Protected Attributes inherited from utils::Function< T >
std::vector< Matrix< T > > saved_tensors_
 

Detailed Description

template<typename T>
class utils::SubFunction< T >

Subtraction function.

Definition at line 164 of file autograd.hpp.

Member Function Documentation

◆ backward()

template<typename T >
std::vector< Matrix< T > > utils::SubFunction< T >::backward ( const Matrix< T > &  grad_output)
inlineoverridevirtual

Backward pass computation.

Parameters
grad_outputGradient from the output
Returns
Gradients with respect to inputs

Implements utils::Function< T >.

Definition at line 170 of file autograd.hpp.

◆ forward()

template<typename T >
Matrix< T > utils::SubFunction< T >::forward ( const std::vector< Variable< T > > &  inputs)
inlineoverridevirtual

Forward pass computation.

Parameters
inputsInput variables
Returns
Output matrix

Implements utils::Function< T >.

Definition at line 166 of file autograd.hpp.


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