checkpoint.hpp File Reference

#include "lbann/callbacks/callback.hpp"
#include "lbann/io/persist.hpp"
#include "lbann/utils/visitor_hooks.hpp"

Include dependency graph for checkpoint.hpp:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
class	lbann::callback::checkpoint
	Checkpoint at given interval in given directory. More...
struct	lbann::callback::checkpoint::header_t< _max_dir_len >

Namespaces
	lbann
	lbann::callback

Enumerations
enum	lbann::callback::callback_phase {   lbann::callback::callback_phase::batch, lbann::callback::callback_phase::epoch, lbann::callback::callback_phase::validation, lbann::callback::callback_phase::inference,   lbann::callback::callback_phase::invalid }

Functions
std::string	lbann::callback::get_trainer_checkpoint_dirname (const std::string &trainer_name, const std::string &dir)
std::string	lbann::callback::get_last_shared_checkpoint_filename (const std::string &alg_name, const std::string &dir)
std::string	lbann::callback::get_last_shared_checkpoint_filename (const std::string &trainer_name, const std::string &alg_name, const std::string &dir)
std::string	lbann::callback::get_shared_checkpoint_dirname (const std::string &alg_name, const std::string &dir, visitor_hook hook, execution_mode mode, size_t epoch, size_t step)
std::string	lbann::callback::get_shared_checkpoint_dirname (const std::string &trainer_name, const std::string &alg_name, const std::string &dir, visitor_hook hook, execution_mode mode, size_t epoch, size_t step)
std::string	lbann::callback::get_last_distributed_checkpoint_filename (const std::string &alg_name, const std::string &dir)
std::string	lbann::callback::get_last_distributed_checkpoint_filename (const std::string &trainer_name, const std::string &alg_name, const std::string &dir)
std::string	lbann::callback::get_distributed_checkpoint_dirname (const std::string &alg_name, const int rank_in_trainer, const std::string &dir, visitor_hook hook, execution_mode mode, size_t epoch, size_t step)
std::string	lbann::callback::get_distributed_checkpoint_dirname (const std::string &trainer_name, const std::string &alg_name, const int rank_in_trainer, const std::string &dir, visitor_hook hook, execution_mode mode, size_t epoch, size_t step)
bool	lbann::callback::write_latest (std::string filename, visitor_hook hook, execution_mode mode, size_t epoch, size_t train)
bool	lbann::callback::read_latest (std::string filename, visitor_hook *hook, execution_mode *mode, size_t *epochLast, size_t *trainLast)
	Reads the "latest" file and returns the epoch number and sample offset for most recent checkpoint. More...
std::unique_ptr< callback_base >	lbann::callback::build_checkpoint_callback_from_pbuf (const google::protobuf::Message &)