JTh L SSKrSSKrSSKrSSKrSSKJr SSKJrJrJ r J r J r J r SSK Jr SSKJrJrJrJr /SQr\ "S5r\ "SS S 9r\\\4r\\S 4r\ "S \\5r"S S\\5r"SS\\\ \5r"SS\\\S 45r"SS\\5r "SS\\5r!"SS\5r""SS\\5r#\4S\\S\\ \$\%4S\ \S\&\#\4Sjjr'g) N)Sequence)castGenericIterableOptionalTypeVarUnion) deprecated)default_generator GeneratorrandpermTensor)DatasetIterableDataset TensorDataset StackDataset ConcatDataset ChainDatasetSubset random_split_T_T_coT) covariant._T_stackc0\rSrSrSrS\4SjrSSjrSrg) r'a}An abstract class representing a :class:`Dataset`. All datasets that represent a map from keys to data samples should subclass it. All subclasses should overwrite :meth:`__getitem__`, supporting fetching a data sample for a given key. Subclasses could also optionally overwrite :meth:`__len__`, which is expected to return the size of the dataset by many :class:`~torch.utils.data.Sampler` implementations and the default options of :class:`~torch.utils.data.DataLoader`. Subclasses could also optionally implement :meth:`__getitems__`, for speedup batched samples loading. This method accepts list of indices of samples of batch and returns list of samples. .. note:: :class:`~torch.utils.data.DataLoader` by default constructs an index sampler that yields integral indices. To make it work with a map-style dataset with non-integral indices/keys, a custom sampler must be provided. returnc[S5e)Nz3Subclasses of Dataset should implement __getitem__.)NotImplementedErrorselfindexs P/var/www/auris/envauris/lib/python3.13/site-packages/torch/utils/data/dataset.py __getitem__Dataset.__getitem__:s!"WXXc[X/5$N)rr!others r#__add__Dataset.__add__Asd]++r&N)r*zDataset[_T_co]rzConcatDataset[_T_co]) __name__ __module__ __qualname____firstlineno____doc__rr$r+__static_attributes__r-r&r#rr's$YEY,r&rc,\rSrSrSrS\\4SjrSrg)rIaAn iterable Dataset. All datasets that represent an iterable of data samples should subclass it. Such form of datasets is particularly useful when data come from a stream. All subclasses should overwrite :meth:`__iter__`, which would return an iterator of samples in this dataset. When a subclass is used with :class:`~torch.utils.data.DataLoader`, each item in the dataset will be yielded from the :class:`~torch.utils.data.DataLoader` iterator. When :attr:`num_workers > 0`, each worker process will have a different copy of the dataset object, so it is often desired to configure each copy independently to avoid having duplicate data returned from the workers. :func:`~torch.utils.data.get_worker_info`, when called in a worker process, returns information about the worker. It can be used in either the dataset's :meth:`__iter__` method or the :class:`~torch.utils.data.DataLoader` 's :attr:`worker_init_fn` option to modify each copy's behavior. Example 1: splitting workload across all workers in :meth:`__iter__`:: >>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_DATALOADER) >>> # xdoctest: +SKIP("Fails on MacOS12") >>> class MyIterableDataset(torch.utils.data.IterableDataset): ... def __init__(self, start, end): ... super(MyIterableDataset).__init__() ... assert end > start, "this example code only works with end >= start" ... self.start = start ... self.end = end ... ... def __iter__(self): ... worker_info = torch.utils.data.get_worker_info() ... if worker_info is None: # single-process data loading, return the full iterator ... iter_start = self.start ... iter_end = self.end ... else: # in a worker process ... # split workload ... per_worker = int(math.ceil((self.end - self.start) / float(worker_info.num_workers))) ... worker_id = worker_info.id ... iter_start = self.start + worker_id * per_worker ... iter_end = min(iter_start + per_worker, self.end) ... return iter(range(iter_start, iter_end)) ... >>> # should give same set of data as range(3, 7), i.e., [3, 4, 5, 6]. >>> ds = MyIterableDataset(start=3, end=7) >>> # Single-process loading >>> print(list(torch.utils.data.DataLoader(ds, num_workers=0))) [tensor([3]), tensor([4]), tensor([5]), tensor([6])] >>> # xdoctest: +REQUIRES(POSIX) >>> # Multi-process loading with two worker processes >>> # Worker 0 fetched [3, 4]. Worker 1 fetched [5, 6]. >>> # xdoctest: +IGNORE_WANT("non deterministic") >>> print(list(torch.utils.data.DataLoader(ds, num_workers=2))) [tensor([3]), tensor([5]), tensor([4]), tensor([6])] >>> # With even more workers >>> # xdoctest: +IGNORE_WANT("non deterministic") >>> print(list(torch.utils.data.DataLoader(ds, num_workers=12))) [tensor([3]), tensor([5]), tensor([4]), tensor([6])] Example 2: splitting workload across all workers using :attr:`worker_init_fn`:: >>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_DATALOADER) >>> class MyIterableDataset(torch.utils.data.IterableDataset): ... def __init__(self, start, end): ... super(MyIterableDataset).__init__() ... assert end > start, "this example code only works with end >= start" ... self.start = start ... self.end = end ... ... def __iter__(self): ... return iter(range(self.start, self.end)) ... >>> # should give same set of data as range(3, 7), i.e., [3, 4, 5, 6]. >>> ds = MyIterableDataset(start=3, end=7) >>> # Single-process loading >>> print(list(torch.utils.data.DataLoader(ds, num_workers=0))) [3, 4, 5, 6] >>> >>> # Directly doing multi-process loading yields duplicate data >>> print(list(torch.utils.data.DataLoader(ds, num_workers=2))) [3, 3, 4, 4, 5, 5, 6, 6] >>> # Define a `worker_init_fn` that configures each dataset copy differently >>> def worker_init_fn(worker_id): ... worker_info = torch.utils.data.get_worker_info() ... dataset = worker_info.dataset # the dataset copy in this worker process ... overall_start = dataset.start ... overall_end = dataset.end ... # configure the dataset to only process the split workload ... per_worker = int(math.ceil((overall_end - overall_start) / float(worker_info.num_workers))) ... worker_id = worker_info.id ... dataset.start = overall_start + worker_id * per_worker ... dataset.end = min(dataset.start + per_worker, overall_end) ... >>> # Mult-process loading with the custom `worker_init_fn` >>> # Worker 0 fetched [3, 4]. Worker 1 fetched [5, 6]. >>> print(list(torch.utils.data.DataLoader(ds, num_workers=2, worker_init_fn=worker_init_fn))) [3, 5, 4, 6] >>> # With even more workers >>> print(list(torch.utils.data.DataLoader(ds, num_workers=12, worker_init_fn=worker_init_fn))) [3, 4, 5, 6] r*c[X/5$r()rr)s r#r+IterableDataset.__add__sTM**r&r-N) r.r/r0r1r2rrr+r3r-r&r#rrIsjX+WU^+r&rcL\rSrSr%Sr\\S4\S'S\SS4SjrSr S r S r g) rzDataset wrapping tensors. Each sample will be retrieved by indexing tensors along the first dimension. Args: *tensors (Tensor): tensors that have the same size of the first dimension. .tensorsrNcV^[U4SjT55(dS5eTUlg)Nc3p># UH+nTSRS5URS5:Hv M- g7f)rN)size).0tensorr:s r# )TensorDataset.__init__..s. ?FVGAJOOA &++a. 0ws36zSize mismatch between tensors)allr:)r!r:s `r#__init__TensorDataset.__init__s8 ?F    + * +  r&cB^[U4SjUR55$)Nc3,># UH oTv M g7fr(r-)r>r?r"s r#r@,TensorDataset.__getitem__..s>vE])tupler:r s `r#r$TensorDataset.__getitem__s>>>>r&c>URSRS5$Nr)r:r=r!s r#__len__TensorDataset.__len__s||A##A&&r&)r:) r.r/r0r1r2rIr__annotations__rCr$rNr3r-r&r#rrs563; D ?'r&rcj\rSrSr%Sr\\\4\S'S\ \ S\ \ SS4Sjr S r S \ 4S jrS rS rg)raMDataset as a stacking of multiple datasets. This class is useful to assemble different parts of complex input data, given as datasets. Example: >>> # xdoctest: +SKIP >>> images = ImageDataset() >>> texts = TextDataset() >>> tuple_stack = StackDataset(images, texts) >>> tuple_stack[0] == (images[0], texts[0]) >>> dict_stack = StackDataset(image=images, text=texts) >>> dict_stack[0] == {'image': images[0], 'text': texts[0]} Args: *args (Dataset): Datasets for stacking returned as tuple. **kwargs (Dataset): Datasets for stacking returned as dict. datasetsargskwargsrNc^U(aRU(a [S5e[US5Tl[U4SjU55(a [S5eUTlgU(aY[ UR 55n[US5Tl[U4SjU55(a [S5eUTlg[S5e)NztSupported either ``tuple``- (via ``args``) or``dict``- (via ``kwargs``) like input/output, but both types are given.rc3T># UHnTR[U5:gv M g7fr(_lengthlenr>datasetr!s r#r@(StackDataset.__init__..sDtG4<<3w</t%(zSize mismatch between datasetsc3T># UHnTR[U5:gv M g7fr(rXr[s r#r@r]sCsG4<<3w</sr^z%At least one dataset should be passed) ValueErrorrZrYanyrSlistvalues)r!rTrUtmps` r#rCStackDataset.__init__s  ^tAw# UH oTv M g7fr(r-)r>r\r"s r#r@+StackDataset.__getitem__..sA=U^=rH) isinstancerSdictitemsrI)r!r"kr\s ` r#r$StackDataset.__getitem__s\ dmmT * *8< 8K8K8MN8M*!Au~%8MN NA4==AAAOsA2indicesc [UR[5(aUVs/sHn0PM nnURR5HupE[ [ USS55(afUR U5n[U5[U5:wa#[S[U5S[U535e[Xc5H upxXxU'M M[X5H upXYX'M M U$UVs/sHn/PM n nURHn[ [ USS55(asUR U5n[U5[U5:wa#[S[U5S[U535e[Xj5Hup{U RU5 M M[X5HupU RXY5 M M U V s/sHn [U 5PM n n U $s snfs snfs sn f)N __getitems__z0Nested dataset's output size mismatch. Expected z, got ) rirSrjrkcallablegetattrrprZr`zipappendrI)r!rn_ dict_batchrlr\rkdatad_sampleidx list_batcht_samplesample tuple_batchs r#rpStackDataset.__getitems__s dmmT * *5<(=WWJ(="mm113 GG^TBCC#009E5zS\1()),WfSZLJ+.e*@&* +A*-W)A &-l *B4 /6!6g"g !6}}G>??,,W5u:W-$%%(\N&U F'*%&%DN&N:uV}: &NA)>""7'Os G)6 G.G3cUR$r()rYrMs r#rNStackDataset.__len__(s ||r&)rYrS)r.r/r0r1r2r rIrjrPrrrCr$rbrprNr3r-r&r#rrsV$E4K  FgenFF4F(B #D#Jr&rc^\rSrSr%Sr\\\\S'\\ \S'\ S5r S\ \SS4U4Sjjr S rS r\\"S \S 9S 55rSrU=r$)ri,zDataset as a concatenation of multiple datasets. This class is useful to assemble different existing datasets. Args: datasets (sequence): List of datasets to be concatenated rScumulative_sizescb/Sp!UH%n[U5nURXB-5 X$- nM' U$rL)rZrt)sequencersels r#cumsumConcatDataset.cumsum8s711AAA HHQUO FAr&rNc >[TU]5 [U5Ul[ UR5S:dS5eURHn[ U[ 5(dMS5e URUR5Ulg)Nrz(datasets should not be an empty iterablez.ConcatDataset does not support IterableDataset) superrCrbrSrZrirrr)r!rSd __class__s r#rCConcatDataset.__init__As X 4==!A%Q'QQ%A!? @? @!% DMM :r&c URS$)NrrMs r#rNConcatDataset.__len__Ks$$R((r&cUS:a)U*[U5:a [S5e[U5U-n[R"URU5nUS:XaUnOXRUS- - nUR UU$)Nrz8absolute value of index should not exceed dataset length)rZr`bisect bisect_rightrrS)r!ry dataset_idx sample_idxs r#r$ConcatDataset.__getitem__Ns 7tc$i Nd)c/C))$*?*?E ! J44[1_EEJ}}[)*55r&z>`cummulative_sizes` attribute is renamed to `cumulative_sizes`)categorycUR$r(rrMs r#cummulative_sizesConcatDataset.cummulative_sizes\s $$$r&)rrS)r.r/r0r1r2rbrrrPint staticmethodrrrCrNr$propertyr FutureWarningrr3 __classcell__rs@r#rr,s~75>""3i;'!2;t;) 6H%  %r&rcJ^\rSrSrSrS\\SS4U4SjjrSrSr S r U=r $) rieaGDataset for chaining multiple :class:`IterableDataset` s. This class is useful to assemble different existing dataset streams. The chaining operation is done on-the-fly, so concatenating large-scale datasets with this class will be efficient. Args: datasets (iterable of IterableDataset): datasets to be chained together rSrNc.>[TU]5 Xlgr()rrCrS)r!rSrs r#rCChainDataset.__init__ps  r&c## URH)n[U[5(dS5eUShvN M+ gN 7f)N*ChainDataset only supports IterableDataset)rSrir)r!rs r#__iter__ChainDataset.__iter__tsEA? <; <LL  s 1?= ?cSnURH-n[U[5(dS5eU[U5- nM/ U$)Nrr)rSrirrZ)r!totalrs r#rNChainDataset.__len__{sOA? <; < SVOE   r&)rS) r.r/r0r1r2rrrCrrNr3rrs@r#rres/!'!2!t!r&rc\rSrSr%Sr\\\S'\\ \S'S\\S\\ SS4Sjr Sr S\ \ S\ \4S jr S rS rg) riz Subset of a dataset at specified indices. Args: dataset (Dataset): The whole Dataset indices (sequence): Indices in the whole set selected for subset r\rnrNcXlX lgr(r\rn)r!r\rns r#rCSubset.__init__s   r&c[U[5(a,URUVs/sHo RUPM sn$URURU$s snfr()rirbr\rn)r!ryis r#r$Subset.__getitem__sP c4 <<# >#Qa# >? ?||DLL-..!?sAc([[URSS55(a8URRUVs/sHo RUPM sn5$UVs/sHo RURUPM sn$s snfs snf)Nrp)rqrrr\rprn)r!rnrys r#rpSubset.__getitems__su GDLL.$? @ @<<,,7-S7Cll3.?7-ST T?FGwLLc!23wG G.TGs B "%Bc,[UR5$r()rZrnrMs r#rNSubset.__len__s4<<  r&r)r.r/r0r1r2rrrPrrrCr$rbrprNr3r-r&r#rrsgU^ c]#4/ HDIH$u+H!r&rr\lengths generatorrc [R"[U5S5(a[U5S::a/n[U5H\upEUS:dUS:a[ SUS35e[ [R "[U5U-55nURU5 M^ [U5[U5- n[U5HnU[U5-nX8==S- ss'M Un[U5H'upIU S:XdM [R"SUS35 M) [U5[U5:wa [ S5e[[U5US9R5n [[[ U5n[!["R$"U5U5V V s/sHup['X X- U 5PM sn n $s sn n f) a Randomly split a dataset into non-overlapping new datasets of given lengths. If a list of fractions that sum up to 1 is given, the lengths will be computed automatically as floor(frac * len(dataset)) for each fraction provided. After computing the lengths, if there are any remainders, 1 count will be distributed in round-robin fashion to the lengths until there are no remainders left. Optionally fix the generator for reproducible results, e.g.: Example: >>> # xdoctest: +SKIP >>> generator1 = torch.Generator().manual_seed(42) >>> generator2 = torch.Generator().manual_seed(42) >>> random_split(range(10), [3, 7], generator=generator1) >>> random_split(range(30), [0.3, 0.3, 0.4], generator=generator2) Args: dataset (Dataset): Dataset to be split lengths (sequence): lengths or fractions of splits to be produced generator (Generator): Generator used for the random permutation. rrzFraction at index z is not between 0 and 1zLength of split at index z- is 0. This might result in an empty dataset.zDSum of input lengths does not equal the length of the input dataset!)r)mathisclosesum enumerater`rfloorrZrtrangewarningswarnr tolistrrrs itertools accumulater) r\rrsubset_lengthsrfracn_items_in_split remainder idx_to_add_atlengthrnoffsets r#rrs< ||CL!$$W):$& )GAax4!8 #5aS8O!PQQ" 3w<$./    ! !"2 3 *L3~#66 y!AN 33M  )Q . )"!"7+IA{ /s3=>, 7|s7|# R  s7|y9@@BG8C='*G")"6"6w"?I INF w&9:I  sF;)(rrrrcollections.abcrtypingrrrrrr typing_extensionsr torchr r r r__all__rrrjstr_T_dictrI_T_tuplerrrrrrrrrfloatrbrr-r&r#rsF  $ED(A@  T]4( sEz    :x 1,gen,Dn+genhuon+h'GE&#+./'0T78$Tn6%GEN6%r?@!WU^!H&7? R[? eCJ' (? "? &* ?r&