torch::data::StatelessDataLoader< Dataset, Sampler > Class Template Reference

A dataloader for stateless datasets. More...

#include <stateless.h>

Inheritance diagram for torch::data::StatelessDataLoader< Dataset, Sampler >:

Public Types
using	super = DataLoaderBase< Dataset, typename Dataset::BatchType, typename Sampler::BatchRequestType >
using	BatchRequestType = BatchRequest
Public Types inherited from torch::data::DataLoaderBase< Dataset, Dataset::BatchType, Sampler::BatchRequestType >
using	BatchType = Dataset::BatchType
using	BatchRequestType = Sampler::BatchRequestType

Public Member Functions
	StatelessDataLoader (Dataset dataset, Sampler sampler, DataLoaderOptions options)
	Constructs the StatelessDataLoader from a dataset, a sampler and some options. More...
Public Member Functions inherited from torch::data::DataLoaderBase< Dataset, Dataset::BatchType, Sampler::BatchRequestType >
	DataLoaderBase (DataLoaderOptions options, std::unique_ptr< Dataset > main_thread_dataset=nullptr)
	Constructs a new DataLoader from a dataset to sample from, options to configure the DataLoader with, and a sampler that specifies the sampling strategy. More...
Iterator< Dataset::BatchType >	begin ()
	Returns an iterator into the DataLoader. More...
Iterator< Dataset::BatchType >	end ()
	Returns a special "sentinel" iterator that compares equal with a non-sentinel iterator once the DataLoader is exhausted. More...
void	join ()
	Joins the DataLoader's worker threads and drains internal queues. More...
const FullDataLoaderOptions &	options () const noexcept
	Returns the options with which the DataLoader was configured.

Additional Inherited Members
Protected Member Functions inherited from torch::data::DataLoaderBase< Dataset, Dataset::BatchType, Sampler::BatchRequestType >
void	prefetch (size_t requested_jobs)
	Schedules requested_jobs many new batches to be fetched. More...
void	prefetch ()
	Schedules the maximum number of jobs (based on the max_jobs option).
optional< BatchType >	next ()
	Returns the next batch of data, or an empty optional if the DataLoader is exhausted. More...
void	worker_thread (Dataset &dataset)
	The function that worker threads run.
void	push_job (T value)
	Convenience method that calls shuttle_.push_job() with the next sequence number. More...
optional< Result >	pop_result ()
	Convenience method that gets the next result from the sequencer.
std::unique_ptr< detail::sequencers::Sequencer< Result > >	new_sequencer ()
	Convenience method that creates a new sequencer based on the enforce_ordering option. More...
Protected Attributes inherited from torch::data::DataLoaderBase< Dataset, Dataset::BatchType, Sampler::BatchRequestType >
const FullDataLoaderOptions	options_
	The options the DataLoader was configured with.
std::unique_ptr< Dataset >	main_thread_dataset_
	The dataset for the main thread, only has a value if the number of worker threads was configured as zero, meaning the main thread has to do all the work (synchronously). More...
size_t	sequence_number_
	The sequence number for the next batch to be retrieved from the dataset. More...
std::vector< std::thread >	workers_
	The worker threads, running the worker_thread() method.
detail::DataShuttle< Job, Result >	shuttle_
	The DataShuttle which takes care of the life cycle of a job.
std::unique_ptr< detail::sequencers::Sequencer< Result > >	sequencer_
	The Sequencer, which handles optional ordering of batches.
bool	joined_
	True if the DataLoader has joined its worker threads.

Detailed Description

template<typename Dataset, typename Sampler>
class torch::data::StatelessDataLoader< Dataset, Sampler >

A dataloader for stateless datasets.

This dataloader follows the traditional PyTorch dataloader design, whereby a (posssibly) stateful sampler produces batch requests for a stateless dataset, which acts as a simple batch request to batch mapping. The batch request will often be an array of indices, and if the dataset is a simple image dataset, the dataset would produce the images at those indices.

Definition at line 25 of file stateless.h.

Constructor & Destructor Documentation

template<typename Dataset , typename Sampler >

torch::data::StatelessDataLoader< Dataset, Sampler >::StatelessDataLoader ( Dataset  dataset, Sampler  sampler, DataLoaderOptions  options  )

inline

Constructs the StatelessDataLoader from a dataset, a sampler and some options.

Definition at line 38 of file stateless.h.

---

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

• torch/csrc/api/include/torch/data/dataloader/stateless.h