.rst .pdf

mlreco.iotools.samplers module

class mlreco.iotools.samplers.AbstractBatchSampler(data_size, minibatch_size, seed=0)

Bases: Generic[torch.utils.data.sampler.T_co]

Samplers that inherit from this class should work out of the box. Just define the __iter__ function __init__ defines self._data_size and self._minibatch_size as well as self._random RNG if needed

__init__(data_size, minibatch_size, seed=0)

Initialize self. See help(type(self)) for accurate signature.

__len__()

__module__ = 'mlreco.iotools.samplers'

__parameters__ = ()

class mlreco.iotools.samplers.RandomSequenceSampler(data_size, minibatch_size, seed=0)

Bases: Generic[torch.utils.data.sampler.T_co]

__iter__()

static create(ds, cfg)

__module__ = 'mlreco.iotools.samplers'

__parameters__ = ()

class mlreco.iotools.samplers.SequentialBatchSampler(data_size, minibatch_size, seed=0)

Bases: Generic[torch.utils.data.sampler.T_co]

__iter__()

static create(ds, cfg)

__module__ = 'mlreco.iotools.samplers'

__parameters__ = ()

class mlreco.iotools.samplers.BootstrapBatchSampler(data_size, minibatch_size, seed=0)

Bases: Generic[torch.utils.data.sampler.T_co]

Sampler used for bootstrap sampling of the entire dataset.

This is particularly useful for training an ensemble of networks (bagging)

__iter__()

static create(ds, cfg)

__module__ = 'mlreco.iotools.samplers'

__parameters__ = ()

mlreco.iotools.factories module mlreco.models package