"Ei1SSKrSSKJrJrJrJr SSKJrJr SSK r /SQr \"SSS9r "SS \\ 5r "S S \ \ 5r"S S \ \ 5r"SS\ \ 5r"SS\ \ 5r"SS\ \\ 5rg)N)IterableIteratorSequenceSized)GenericTypeVar) BatchSampler RandomSamplerSamplerSequentialSamplerSubsetRandomSamplerWeightedRandomSampler_T_coT) covariantc,\rSrSrSrS\\4SjrSrg)r aBase class for all Samplers. Every Sampler subclass has to provide an :meth:`__iter__` method, providing a way to iterate over indices or lists of indices (batches) of dataset elements, and may provide a :meth:`__len__` method that returns the length of the returned iterators. Example: >>> # xdoctest: +SKIP >>> class AccedingSequenceLengthSampler(Sampler[int]): >>> def __init__(self, data: List[str]) -> None: >>> self.data = data >>> >>> def __len__(self) -> int: >>> return len(self.data) >>> >>> def __iter__(self) -> Iterator[int]: >>> sizes = torch.tensor([len(x) for x in self.data]) >>> yield from torch.argsort(sizes).tolist() >>> >>> class AccedingSequenceLengthBatchSampler(Sampler[List[int]]): >>> def __init__(self, data: List[str], batch_size: int) -> None: >>> self.data = data >>> self.batch_size = batch_size >>> >>> def __len__(self) -> int: >>> return (len(self.data) + self.batch_size - 1) // self.batch_size >>> >>> def __iter__(self) -> Iterator[List[int]]: >>> sizes = torch.tensor([len(x) for x in self.data]) >>> for batch in torch.chunk(torch.argsort(sizes), len(self)): >>> yield batch.tolist() .. note:: The :meth:`__len__` method isn't strictly required by :class:`~torch.utils.data.DataLoader`, but is expected in any calculation involving the length of a :class:`~torch.utils.data.DataLoader`. returnc[eN)NotImplementedErrorselfs U/home/mande/repo/quber/.venv/lib/python3.13/site-packages/torch/utils/data/sampler.py__iter__Sampler.__iter__Bs!!N) __name__ __module__ __qualname____firstlineno____doc__rrr__static_attributes__rrrr r s#J"(5/"rr cX\rSrSr%Sr\\S'S\SS4SjrS\\ 4Sjr S\ 4Sjr S r g) r azSamples elements sequentially, always in the same order. Args: data_source (Sized): data source to sample from. Must implement __len__. data_sourcerNcXlgrr&)rr&s r__init__SequentialSampler.__init__js&rcP[[[UR555$r)iterrangelenr&rs rrSequentialSampler.__iter__msE#d../011rc,[UR5$r)r.r&rs r__len__SequentialSampler.__len__ps4##$$rr() rrr r!r"r__annotations__r)rintrr1r#rrrr r as> 'E'd'2(3-2%%rr c \rSrSr%Sr\\S'\\S'S S\S\S\S-SS4Sjjr \ S\4S j5r S\ \4S jr S\4S jrS rg)r taSamples elements randomly. If without replacement, then sample from a shuffled dataset. If with replacement, then user can specify :attr:`num_samples` to draw. Args: data_source (Sized): data source to sample from. Must implement __len__. replacement (bool): samples are drawn on-demand with replacement if ``True``, default=``False`` num_samples (int): number of samples to draw, default=`len(dataset)`. generator (Generator): Generator used in sampling. r& replacementN num_samplesrc0XlX lX0lX@l[ UR[ 5(d[ SUR35e[ UR[5(aURS::a[SUR35eg)N;replacement should be a boolean value, but got replacement=rDnum_samples should be a positive integer value, but got num_samples=) r&r7 _num_samples generator isinstancebool TypeErrorr8r4 ValueError)rr&r7r8r=s rr)RandomSampler.__init__s'&'"$**D11MdN^N^M_` $**C00D4D4D4IVW[WgWgVhi 5Jrc^URc[UR5$UR$r)r<r.r&rs rr8RandomSampler.num_sampless-    $t''( (   rc## [UR5nURco[[R "S[R S9R5R55n[R"5nURU5 O URnUR(a[URS-5H>n[R"US[R US9R5ShvN M@ [R"UURS-4[R US9R5ShvN g[URU-5H-n[R "XS9R5ShvN M/ [R "XS9R5SURU-ShvN gNNNGN 7f)Nrdtype )rH)highsizerGr=r=)r.r&r=r4torchemptyint64random_item Generator manual_seedr7r-r8randinttolistrandperm)rnseedr=_s rrRandomSampler.__iter__sy   ! >> !u{{2U[[9AACHHJKD)I  ! !$ 'I   4++r12 ==ekkY&(3}}&&+-kk#  fh   4++q01 >>!AHHJJJ2~~a=DDF&$""Q&   K sJDG!GA G!GAG!G=G!GG!G!G!G!cUR$rr8rs rr1RandomSampler.__len__r)r<r&r=r7)FNN)rrr r!r"rr3r?r4r)propertyr8rrr1r#rrrr r ts  ""& 4Z   ,!S!! (3-6  rr ch\rSrSr%Sr\\\S'S S\\SS4SjjrS\ \4Sjr S\4Sjr S r g) r zSamples elements randomly from a given list of indices, without replacement. Args: indices (sequence): a sequence of indices generator (Generator): Generator used in sampling. indicesNrcXlX lgr)rar=)rrar=s rr)SubsetRandomSampler.__init__s  "rc## [R"[UR5URS9R 5HnURUv M g7fNrK)rLrUr.rar=rT)ris rrSubsetRandomSampler.__iter__s?DLL 1T^^LSSUA,,q/ !VsAAc,[UR5$r)r.rars rr1SubsetRandomSampler.__len__s4<<  r)r=rar) rrr r!r"rr4r3r)rrr1r#rrrr r sFc]# #$#"(3-"!!rr c \rSrSr%Sr\R \S'\\S'\ \S'S S\ \ S\S\ SS4Sjjr S\ \4S jrS\4S jrS rg) ra~Samples elements from ``[0,..,len(weights)-1]`` with given probabilities (weights). Args: weights (sequence) : a sequence of weights, not necessary summing up to one num_samples (int): number of samples to draw replacement (bool): if ``True``, samples are drawn with replacement. If not, they are drawn without replacement, which means that when a sample index is drawn for a row, it cannot be drawn again for that row. generator (Generator): Generator used in sampling. Example: >>> # xdoctest: +IGNORE_WANT("non-deterministic") >>> list( ... WeightedRandomSampler( ... [0.1, 0.9, 0.4, 0.7, 3.0, 0.6], 5, replacement=True ... ) ... ) [4, 4, 1, 4, 5] >>> list( ... WeightedRandomSampler( ... [0.9, 0.4, 0.05, 0.2, 0.3, 0.1], 5, replacement=False ... ) ... ) [0, 1, 4, 3, 2] weightsr8r7Nrc[U[5(a[U[5(dUS::a[SU35e[U[5(d[SU35e[R "U[R S9n[UR5S:wa![S[UR535eXPl X l X0l X@l g)Nrr;r:rFz=weights should be a 1d sequence but given weights have shape )r>r4r?rArL as_tensordoubler.shapetuplerlr8r7r=)rrlr8r7r=weights_tensors rr)WeightedRandomSampler.__init__s;,,+t,,aVWbVcd +t,,Mk][  E ~## $ )&&+N,@,@&A%BD  & &&"rc## [R"URURURUR S9n[ UR55ShvN gN7fre)rL multinomialrlr8r7r=r,rT)r rand_tensors rrWeightedRandomSampler.__iter__sK'' LL$**D,<,<  **,---sAA&A$A&cUR$rr[rs rr1WeightedRandomSampler.__len__r]r)r=r8r7rl)TN)rrr r!r"rLTensorr3r4r?rfloatr)rrr1r#rrrrrss4\\ ! #%## #  #@.(3-.   rrcl\rSrSrSrS\\\\-S\S\SS4Sjr S\ \ \4S jr S\4S jr S rg) r iaWraps another sampler to yield a mini-batch of indices. Args: sampler (Sampler or Iterable): Base sampler. Can be any iterable object batch_size (int): Size of mini-batch. drop_last (bool): If ``True``, the sampler will drop the last batch if its size would be less than ``batch_size`` Example: >>> list( ... BatchSampler( ... SequentialSampler(range(10)), batch_size=3, drop_last=False ... ) ... ) [[0, 1, 2], [3, 4, 5], [6, 7, 8], [9]] >>> list( ... BatchSampler(SequentialSampler(range(10)), batch_size=3, drop_last=True) ... ) [[0, 1, 2], [3, 4, 5], [6, 7, 8]] sampler batch_size drop_lastrNc[U[5(a[U[5(dUS::a[SU35e[U[5(d[SU35eXlX lX0lg)NrzBbatch_size should be a positive integer value, but got batch_size=z7drop_last should be a boolean value, but got drop_last=)r>r4r?rAr~rr)rr~rrs rr)BatchSampler.__init__4su:s++*d++QTU_T`a )T**I)U  $"rc#\# [UR5nUR(a)U/UR-n[ USS06H n/UQv M g/[ R "XR5QnU(a0Uv /[ R "XR5QnU(aM/gg7f)NstrictF)r,r~rrzip itertoolsislice)r sampler_iterargsbatch_droplastbatchs rrBatchSampler.__iter__MsDLL) >> >DOO3D"%t":E":'''#;Gi&&|__EFE J)**<IJ%s B&B,*B,cUR(a"[UR5UR-$[UR5UR-S- UR-$)Nrn)rr.r~rrs rr1BatchSampler.__len__ZsI >>t||$7 7 %7!;O Or)rrr~)rrr r!r"r r4rr?r)rlistrr1r#rrrr r sd*# -## #  #2 K(49- KPPrr )rcollections.abcrrrrtypingrrrL__all__rr r4r r r rrr rrrrs??#   4('"gen'"J% %&H GCLH V!'#,!,F GCLF RDP749%DPr