pulsatrix::ShuffleSampler Class Reference

Visits indices [0, dataset_size) in a seeded pseudo-random permutation – reproducible across runs given the same seed (std::mt19937), re-shuffled fresh each reset() call (a new epoch is a new permutation, not the same one repeated). More...

#include <sampler.hpp>

Inheritance diagram for pulsatrix::ShuffleSampler:

Collaboration diagram for pulsatrix::ShuffleSampler:

Public Member Functions
	ShuffleSampler (unsigned seed)
void	reset (int64_t dataset_size) override
	Begins a new epoch over a dataset of the given size.
std::optional< int64_t >	next () override
	Fetches the next index in this epoch's order.
Public Member Functions inherited from pulsatrix::Sampler
virtual	~Sampler ()=default

Detailed Description

Visits indices [0, dataset_size) in a seeded pseudo-random permutation – reproducible across runs given the same seed (std::mt19937), re-shuffled fresh each reset() call (a new epoch is a new permutation, not the same one repeated).

Constructor & Destructor Documentation

ShuffleSampler()

pulsatrix::ShuffleSampler::ShuffleSampler ( unsigned  seed )

explicit

Member Function Documentation

next()

std::optional< int64_t > pulsatrix::ShuffleSampler::next ( )

overridevirtual

Fetches the next index in this epoch's order.

Returns

The next index, or std::nullopt once every index has been produced.

Implements pulsatrix::Sampler.

reset()

void pulsatrix::ShuffleSampler::reset ( int64_t  dataset_size )

overridevirtual

Begins a new epoch over a dataset of the given size.

Parameters

dataset_size

Number of samples to produce indices for. Must be >= 0 – internal invariant (PULSATRIX_ASSERT, not throw): DataLoader always calls this with Dataset::size()'s return value directly, never unvalidated external input.

Implements pulsatrix::Sampler.

---

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

• include/pulsatrix/sampler.hpp