Χg 3XddlmZmZmZmZmZmZmZmZm Z ddl Z gdZ eddZ Gddee Z Gd d e eZGd d e eZGd de eZGdde eZGdde eeZdS)) GenericIterableIteratorListOptionalSequenceSizedTypeVarUnionN) BatchSampler RandomSamplerSamplerSequentialSamplerSubsetRandomSamplerWeightedRandomSampler_T_coT) covariantcHeZdZdZddeeddfdZdeefdZ dS)ra;Base 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. Args: data_source (Dataset): This argument is not used and will be removed in 2.2.0. You may still have custom implementation that utilizes it. 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`. N data_sourcereturnc@|ddl}|ddSdS)Nrzz`data_source` argument is not used and will be removed in 2.2.0.You may still have custom implementation that utilizes it.)warningswarn)selfrrs T/var/www/html/ai-engine/env/lib/python3.11/site-packages/torch/utils/data/sampler.py__init__zSampler.__init__Hs?  " OOO MMM      # "ctN)NotImplementedErrorrs r__iter__zSampler.__iter__Qs!!rr) __name__ __module__ __qualname____doc__rr rrrr"rrrrsi''RHUOt"(5/""""""rrcReZdZUdZeed<deddfdZdeefdZ defdZ dS)rz~Samples elements sequentially, always in the same order. Args: data_source (Dataset): dataset to sample from rrNc||_dSr)r)rrs rrzSequentialSampler.__init__ys&rc^ttt|jSr)iterrangelenrr!s rr"zSequentialSampler.__iter__|s#E#d.//00111rc*t|jSr)r-rr!s r__len__zSequentialSampler.__len__s4#$$$r) r#r$r%r&r __annotations__rrintr"r/r'rrrrps 'E'd''''2(3-2222%%%%%%%rrc eZdZUdZeed<eed< d dededeeddfdZ e defd Z de efd Z defd ZdS) r aSamples 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 (Dataset): dataset to sample from 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 replacementFN num_samplesrc||_||_||_||_t |jt st d|jt |jtr |jdkrtd|jdS)N;replacement should be a boolean value, but got replacement=rDnum_samples should be a positive integer value, but got num_samples=) rr3 _num_samples generator isinstancebool TypeErrorr4r1 ValueError)rrr3r4r9s rrzRandomSampler.__init__s'&'"$*D11 `dN^`` $*C00 D4D4I4IiW[Wgii 5J4IrcF|jt|jS|jSr)r8r-rr!s rr4zRandomSampler.num_sampless'   $t'(( (  rc#Kt|j}|jztt jdtj}t j }| |n|j}|j rt|j dzD]<}t j|dtj|Ed{V=t j||j dzftj|Ed{VdSt|j |zD]0}t j||Ed{V1t j||d|j |zEd{VdS)Nr'dtype )rB)highsizerAr9r9)r-rr9r1torchemptyint64random_item Generator manual_seedr3r,r4randinttolistrandperm)rnseedr9_s rr"zRandomSampler.__iter__s  ! ! > !u{2U[999AACCHHJJKKD))I  ! !$ ' ' ' 'I   4+r122   =ekY&((}&+-k#  fhh          4+q011 K K >!yAAAHHJJJJJJJJJJ~a9===DDFF&$"Q&&         rc|jSrr4r!s rr/zRandomSampler.__len__ r)FNN)r#r$r%r&r r0r;rr1rpropertyr4rr"r/r'rrr r s   "%) c]   ,!S!!!X! (3-6       rr cleZdZUdZeeed<ddeeddfdZdeefdZ defdZ dS) rzSamples elements randomly from a given list of indices, without replacement. Args: indices (sequence): a sequence of indices generator (Generator): Generator used in sampling. indicesNrc"||_||_dSr)rXr9)rrXr9s rrzSubsetRandomSampler.__init__s "rc#Ktjt|j|jD]}|j|VdSNrE)rFrOr-rXr9)ris rr"zSubsetRandomSampler.__iter__sODL 1 1T^LLL " "A,q/ ! ! ! ! " "rc*t|jSr)r-rXr!s rr/zSubsetRandomSampler.__len__s4<   rr) r#r$r%r&rr1r0rrr"r/r'rrrrsc]## #$####"(3-""""!!!!!!!rrc eZdZUdZejed<eed<eed< d de e dededdfdZ de efd Z defd ZdS) raNSamples 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] weightsr4r3TNrct|trt|ts|dkrtd|t|tstd|t j|tj}t|jdkr$tdt|j||_ ||_ ||_ ||_ dS)Nrr7r6r@z=weights should be a 1d sequence but given weights have shape )r:r1r;r=rF as_tensordoubler-shapetupler_r4r3r9)rr_r4r3r9weights_tensors rrzWeightedRandomSampler.__init__s;,, +t,, adWbdd +t,, [k[[  EEE ~# $ $ ) )D&+N,@&A&ADD  & &&"rc#Ktj|j|j|j|j}t |Ed{VdSr[)rF multinomialr_r4r3r9r+rN)r rand_tensors rr"zWeightedRandomSampler.__iter__sd' L$*D,<    **,,-----------rc|jSrrTr!s rr/zWeightedRandomSampler.__len__!rUr)TN)r#r$r%r&rFTensorr0r1r;rfloatrrr"r/r'rrrrs$\ ! ##%## #  ####@.(3-....        rrceZdZdZdeeeeefdededdfdZ de e efdZ defd Z dS) r aiWraps 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_lastrNct|trt|ts|dkrtd|t|tstd|||_||_||_dS)NrzBbatch_size should be a positive integer value, but got batch_size=z7drop_last should be a boolean value, but got drop_last=)r:r1r;r=rnrorp)rrnrorps rrzBatchSampler.__init__5s:s++ *d++ QaU_aa )T** U)UU  $"rc#VK|jrMt|j fdt|jD}|Vn#t $rYdSwxYw8dg|jz}d}|jD](}|||<|dz }||jkr|Vd}dg|jz})|dkr|d|VdSdS)NTc.g|]}tSr')next).0rR sampler_iters r z)BatchSampler.__iter__..Ts!PPPAT,//PPPrrra)rpr+rnr,ro StopIteration)rbatch idx_in_batchidxrvs @rr"zBatchSampler.__iter__Ns > + --L PPPPt9O9OPPPEKKKK$EE  C$/)EL| 2 2&)l#! 4?22KKK#$LC$/1EaM\M****** s$A AAc|jrt|j|jzSt|j|jzdz |jzS)Nra)rpr-rnror!s rr/zBatchSampler.__len__esI > Pt|$$7 7 %%7!;O Or)r#r$r%r&r rr1rr;rrrr"r/r'rrr r %s  #ws|Xc]23## #  ####2+(49-++++.PPPPPPPrr )typingrrrrrrr r r rF__all__rrr1rr rrr r'rrrs                           4(((4"4"4"4"4"gen4"4"4"d%%%%% %%%&H H H H H GCLH H H V!!!!!'#,!!!,> > > > > GCL> > > BHPHPHPHPHP749%HPHPHPHPHPr