a 7SicJ*@sddlZddlmZddlmZmZmZmZmZmZm Z m Z m Z gdZ edddZ Gdd d e e ZGd d d eeZGd d d eeZGdddeeZGdddeeZGdddeeeZdS)N)Tensor) IteratorIterableOptionalSequenceListTypeVarGenericSizedUnion) BatchSampler RandomSamplerSamplerSequentialSamplerSubsetRandomSamplerWeightedRandomSamplerT_coT) covariantc@s6eZdZdZeeddddZeedddZ dS) raBase class for all Samplers. Every Sampler subclass has to provide an :meth:`__iter__` method, providing a way to iterate over indices of dataset elements, and a :meth:`__len__` method that returns the length of the returned iterators. .. 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_sourcereturncCsdSNselfrrrT/var/www/html/django/DPS/env/lib/python3.9/site-packages/torch/utils/data/sampler.py__init__szSampler.__init__rcCstdSr)NotImplementedErrorrrrr__iter__!szSampler.__iter__) __name__ __module__ __qualname____doc__rr rrrr rrrrrs rc@sJeZdZUdZeed<eddddZeeddd Z edd d Z dS) rz~Samples elements sequentially, always in the same order. Args: data_source (Dataset): dataset to sample from rNrcCs ||_dSr)rrrrrrHszSequentialSampler.__init__rcCsttt|jSr)iterrangelenrrrrrr KszSequentialSampler.__iter__cCs t|jSr)r'rrrrr__len__NszSequentialSampler.__len__) r!r"r#r$r __annotations__rrintr r(rrrrr@s rc@sneZdZUdZeed<eed<deeeeddddZ e ed d d Z e ed d d Z ed ddZdS)r aSamples elements randomly. If without replacement, then sample from a shuffled dataset. If with replacement, then user can specify :attr:`num_samples` to draw. Args: data_source (Dataset): dataset to sample from replacement (bool): samples are drawn on-demand with replacement if ``True``, default=``False`` num_samples (int): number of samples to draw, default=`len(dataset)`. generator (Generator): Generator used in sampling. r replacementFN)rr+ num_samplesrcCs^||_||_||_||_t|jts4td|jt|jt rJ|jdkrZt d|jdS)N=replacement should be a boolean value, but got replacement={}rFnum_samples should be a positive integer value, but got num_samples={}) rr+ _num_samples generator isinstancebool TypeErrorformatr,r* ValueError)rrr+r,r0rrrr_s zRandomSampler.__init__rcCs|jdurt|jS|jSr)r/r'rrrrrr,ns  zRandomSampler.num_samplesccst|j}|jdurDttjdtjd}t }| |n|j}|j rt |j dD]"}tj|dtj|dEdHq^tj||j dftj|dEdHnRt |j |D]}tj||dEdHqtj||dd|j |EdHdS)Nrdtype )r8)highsizer7r0r0)r'rr0r*torchemptyint64random_item Generator manual_seedr+r&r,randinttolistrandperm)rnseedr0_rrrr us    (zRandomSampler.__iter__cCs|jSrr,rrrrr(szRandomSampler.__len__)FNN)r!r"r#r$r r)r2rr*rpropertyr,rr r(rrrrr Rs   r c@sTeZdZUdZeeed<d eeddddZeeddd Z edd d Z 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. indicesN)rKrcCs||_||_dSr)rKr0)rrKr0rrrrszSubsetRandomSampler.__init__rccs,tjt|j|jdD]}|j|VqdSNr;)r<rEr'rKr0)rirrrr szSubsetRandomSampler.__iter__cCs t|jSr)r'rKrrrrr(szSubsetRandomSampler.__len__)N) r!r"r#r$rr*r)rrr r(rrrrrs  rc@sdeZdZUdZeed<eed<eed<dee eedddd Z e ed d d Z ed d dZ dS)raSamples elements from ``[0,..,len(weights)-1]`` with given probabilities (weights). Args: weights (sequence) : a sequence of weights, not necessary summing up to one num_samples (int): number of samples to draw replacement (bool): if ``True``, samples are drawn with replacement. If not, they are drawn without replacement, which means that when a sample index is drawn for a row, it cannot be drawn again for that row. generator (Generator): Generator used in sampling. Example: >>> list(WeightedRandomSampler([0.1, 0.9, 0.4, 0.7, 3.0, 0.6], 5, replacement=True)) [4, 4, 1, 4, 5] >>> list(WeightedRandomSampler([0.9, 0.4, 0.05, 0.2, 0.3, 0.1], 5, replacement=False)) [0, 1, 4, 3, 2] weightsr,r+TN)rNr,r+rcCsjt|trt|ts|dkr*td|t|tsBtd|tj|tjd|_||_ ||_ ||_ dS)Nrr.r-r6) r1r*r2r5r4r< as_tensordoublerNr,r+r0)rrNr,r+r0rrrrs zWeightedRandomSampler.__init__rccs0tj|j|j|j|jd}t|EdHdSrL)r< multinomialrNr,r+r0r%rD)rZ rand_tensorrrrr szWeightedRandomSampler.__iter__cCs|jSrrIrrrrr(szWeightedRandomSampler.__len__)TN)r!r"r#r$rr)r*r2rfloatrrr r(rrrrrs   rc@sXeZdZdZeeeeefeeddddZ e e edddZ edd d 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]] N)sampler batch_size drop_lastrcCsXt|trt|ts|dkr*td|t|tsBtd|||_||_||_dS)NrzDbatch_size should be a positive integer value, but got batch_size={}z9drop_last should be a boolean value, but got drop_last={})r1r*r2r5r4rSrTrU)rrSrTrUrrrrs zBatchSampler.__init__rc#s|jrNt|jz"fddt|jD}|VWqtyHYqYq0qnbdg|j}d}|jD]4}|||<|d7}||jkrd|Vd}dg|j}qd|dkr|d|VdS)Ncsg|] }tqSr)next).0rHZ sampler_iterrr z)BatchSampler.__iter__..r)rUr%rSr&rT StopIteration)rbatchZ idx_in_batchidxrrXrr s$      zBatchSampler.__iter__cCs4|jrt|j|jSt|j|jd|jSdS)Nr[)rUr'rSrTrrrrr(szBatchSampler.__len__)r!r"r#r$r rr*rr2rrrr r(rrrrr s$r )r<rtypingrrrrrrr r r __all__rrr*rr rrr rrrrs , .9+