PhddlZddlZddlmZmZddlmZddlmZddl m Z m Z m Z m Z mZmZmZmZddgZedd ZGd deeZed Gd deeZy)N)SamplerSequentialSampler)functional_datapipe) IterDataPipe)DictIteratorListOptionalSizedTupleTypeTypeVarSamplerIterDataPipeShufflerIterDataPipeT_coT) covariantc eZdZUdZeed<eed<eddfdedeede e de e ddf fd Z de efd Zdefd ZxZS) ra7 Generate sample elements using the provided ``Sampler`` (defaults to :class:`SequentialSampler`). Args: datapipe: IterDataPipe to sample from sampler: Sampler class to generate sample elements from input DataPipe. Default is :class:`SequentialSampler` for IterDataPipe datapipesamplerN sampler_argssampler_kwargsreturnct|tsJdt| ||_|dn||_|in||_||j d|ji|j |_y)Nz;Sampler class requires input datapipe implemented `__len__` data_source) isinstancer super__init__rrrr)selfrrrr __class__s xC:\Users\daisl\Desktop\realtime-object-detection\venv\Lib\site-packages\torch/utils/data/datapipes/iter/combinatorics.pyrzSamplerIterDataPipe.__init__sx (E* J I J*   "."6BL$2$:b 1 1dt}}dPTPcPcd c,t|jSN)iterrrs r!__iter__zSamplerIterDataPipe.__iter__-sDLL!!r"ct|jtrt|jSt t |j dNz# instance doesn't have valid length)rrr len TypeErrortype__name__r&s r!__len__zSamplerIterDataPipe.__len__0s= dllE *t||$ $4:..//RSTTr")r- __module__ __qualname____doc__r__annotations__rrr r r rrrrr'intr. __classcell__r s@r!rrs +<1526 e' ew- e ( e"*$ e  e"(4."UUr"shufflec eZdZUdZeeed<eed<eeed<e ed<e eed<e jed<dd d deeded ed d ffdZ ddZdefdZd eefdZd efdZddZdZdZdZxZS)ra Shuffle the input DataPipe with a buffer (functional name: ``shuffle``). The buffer with ``buffer_size`` is filled with elements from the datapipe first. Then, each item will be yielded from the buffer by reservoir sampling via iterator. ``buffer_size`` is required to be larger than ``0``. For ``buffer_size == 1``, the datapipe is not shuffled. In order to fully shuffle all elements from datapipe, ``buffer_size`` is required to be greater than or equal to the size of datapipe. When it is used with :class:`torch.utils.data.DataLoader`, the methods to set up random seed are different based on :attr:`num_workers`. For single-process mode (:attr:`num_workers == 0`), the random seed is set before the :class:`~torch.utils.data.DataLoader` in the main process. For multi-process mode (:attr:`num_worker > 0`), `worker_init_fn` is used to set up a random seed for each worker process. Args: datapipe: The IterDataPipe being shuffled buffer_size: The buffer size for shuffling (default to ``10000``) unbatch_level: Specifies if it is necessary to unbatch source data before applying the shuffle Example: >>> # xdoctest: +SKIP >>> from torchdata.datapipes.iter import IterableWrapper >>> dp = IterableWrapper(range(10)) >>> shuffle_dp = dp.shuffle() >>> list(shuffle_dp) [0, 4, 1, 6, 3, 2, 9, 5, 7, 8] r buffer_size_buffer_enabled_seed_rngi'r)r8 unbatch_levelr=rNct|g|_|dkDsJd|dk(r||_n|j ||_||_d|_d|_tj|_ y)Nrz#buffer_size should be larger than 0)r=T) rrr9runbatchr8r:r;randomRandomr<)rrr8r=r s r!rzShufflerIterDataPipe.__init__asr $& QE EE A $DM$,,=,IDM&  MMO r"c||_|Sr$)r:)rr6s r! set_shufflez ShufflerIterDataPipe.set_shuffleus  r"seedc||_|Sr$)r;)rrDs r!set_seedzShufflerIterDataPipe.set_seedys  r"c#nK|js|jEd{y|jD]}t|j|jk(rV|j j dt|jdz }|j||c}|j|<|{|jj||jr]|j j dt|jdz }|jj||jr\yy7w)Nr) r:rr*r9r8r<randintappendpop)rxidxvals r!r'zShufflerIterDataPipe.__iter__}s}}}} $ $]]t||$(8(88))++As4<*Cc*ILL''* #,,ii''3t||+rlsv 7E<NNN  v&#U,t,#ULY<- r"