torch::optim::detail::OptimizerBase Class Reference

Base class for all optimizers, that does not yet define a step() mechanism. More...

#include <optimizer.h>

Inheritance diagram for torch::optim::detail::OptimizerBase:

Public Member Functions
	OptimizerBase (std::vector< Tensor > parameters)
	Constructs the Optimizer from a vector of parameters.
void	add_parameters (const std::vector< Tensor > &parameters)
	Adds the given vector of parameters to the optimizer's parameter list.
virtual void	zero_grad ()
	Zeros out the gradients of all parameters.
const std::vector< Tensor > &	parameters () const noexcept
	Provides a const reference to the parameters this optimizer holds.
std::vector< Tensor > &	parameters () noexcept
	Provides a reference to the parameters this optimizer holds.
size_t	size () const noexcept
	Returns the number of parameters referenced by the optimizer.
virtual void	save (serialize::OutputArchive &archive) const
	Serializes the optimizer state into the given archive.
virtual void	load (serialize::InputArchive &archive)
	Deserializes the optimizer state from the given archive.

Protected Member Functions
template<typename T >
T &	buffer_at (std::vector< T > &buffers, size_t index)
	Accesses a buffer at the given index. More...
Tensor &	buffer_at (std::vector< Tensor > &buffers, size_t index)
	Accesses a buffer at the given index, converts it to the type of the parameter at the corresponding index (a no-op if they match). More...

Protected Attributes
std::vector< Tensor >	parameters_
	The parameters this optimizer optimizes.

Detailed Description

Base class for all optimizers, that does not yet define a step() mechanism.

All it specifies is that optimizers must be supplied with a vector of parameters. It also defines certain methods that all optimizers shall have, such as zero_grad.

Definition at line 34 of file optimizer.h.

Member Function Documentation

template<typename T >

T& torch::optim::detail::OptimizerBase::buffer_at ( std::vector< T > &  buffers, size_t  index  )

inlineprotected

Accesses a buffer at the given index.

Additionally, zeros out the buffers when this is called on the index

Definition at line 68 of file optimizer.h.

Tensor & torch::optim::detail::OptimizerBase::buffer_at ( std::vector< Tensor > &  buffers, size_t  index  )

protected

Accesses a buffer at the given index, converts it to the type of the parameter at the corresponding index (a no-op if they match).

Additionally, zeros out the buffers when this is called on the index

Definition at line 43 of file optimizer.cpp.

---

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

• torch/csrc/api/include/torch/optim/optimizer.h
• torch/csrc/api/src/optim/optimizer.cpp