o iB @sddlZddlZddlZddlmZmZmZmZmZm Z m Z m Z m Z m Z ddlmZmZddlmZddlmZmZgdZe dd d Ze d Ze eefZe ed fZe d eeZGdddeeZGdddeeZGdddee ed fZGdddeeZ GdddeeZ!GdddeZ"GdddeeZ#efdeede e e$e%fdeedee#efd d!Z&dS)"N) GenericIterableIteratorListOptionalSequenceTupleTypeVarUnionDict)default_generatorrandperm) _accumulate) GeneratorTensor)DatasetIterableDataset TensorDataset StackDataset ConcatDataset ChainDatasetSubset random_splitT_coT) covariantT.T_stackc@s(eZdZdZdefddZd dd Zd S) raAn 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 a 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. returncCtd)Nz3Subclasses of Dataset should implement __getitem__.NotImplementedErrorselfindexr%g/var/www/html/eduruby.in/lip-sync/lip-sync-env/lib/python3.10/site-packages/torch/utils/data/dataset.py __getitem__<zDataset.__getitem__other Dataset[T_co]ConcatDataset[T_co]cC t||gSN)rr#r)r%r%r&__add__C zDataset.__add__N)r)r*rr+)__name__ __module__ __qualname____doc__rr'r/r%r%r%r&r)src@s4eZdZdZdeefddZdeefddZdS) raHAn 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) >>> # Mult-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] rcCr)Nz8Subclasses of IterableDataset should implement __iter__.r r#r%r%r&__iter__r(zIterableDataset.__iter__r)cCr,r-)rr.r%r%r&r/r0zIterableDataset.__add__N) r1r2r3r4rrr6rr/r%r%r%r&rKskrc@sDeZdZUdZeedfed<deddfddZdd Zd d Z dS) 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. .tensorsrNcs(tfddDsJd|_dS)Nc3s(|]}dd|dkVqdS)rN)size.0Ztensorr7r%r& s&z)TensorDataset.__init__..zSize mismatch between tensors)allr7)r#r7r%r;r&__init__s zTensorDataset.__init__cstfdd|jDS)Nc3|]}|VqdSr-r%r9r$r%r&r<z,TensorDataset.__getitem__..)tupler7r"r%r@r&r'szTensorDataset.__getitem__cCs|jddSNr)r7r8r5r%r%r&__len__szTensorDataset.__len__) r1r2r3r4rr__annotations__r>r'rDr%r%r%r&rs  rc@sPeZdZUdZeeefed<dee dee ddfddZ d d Z d d Z dS) raDataset 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. datasetsargskwargsrNcs|r#|rtdt|d_tfdd|Drtd|_dS|rFt|}t|d_tfdd|DrAtd|_dStd)NztSupported either ``tuple``- (via ``args``) or``dict``- (via ``kwargs``) like input/output, but both types are given.rc3|] }jt|kVqdSr-_lengthlenr:datasetr5r%r&r<z(StackDataset.__init__..zSize mismatch between datasetsc3rIr-rJrMr5r%r&r<rOz%At least one dataset should be passed) ValueErrorrLrKanyrFlistvalues)r#rGrHtmpr%r5r&r>s   zStackDataset.__init__cs<t|jtrfdd|jDStfdd|jDS)Ncsi|] \}}||qSr%r%)r:krNr@r%r& sz,StackDataset.__getitem__..c3r?r-r%rMr@r%r&r<rAz+StackDataset.__getitem__..) isinstancerFdictitemsrBr"r%r@r&r's zStackDataset.__getitem__cCs|jSr-)rKr5r%r%r&rDszStackDataset.__len__) r1r2r3r4r rBrXrErrr>r'rDr%r%r%r&rs  rcsteZdZUdZeeeed<eeed<e ddZ de eddffdd Z d d Z d d ZeddZZS)rzDataset as a concatenation of multiple datasets. This class is useful to assemble different existing datasets. Args: datasets (sequence): List of datasets to be concatenated rFcumulative_sizescCs6gd}}|D]}t|}|||||7}q|SrC)rLappend)sequencerselr%r%r&cumsums  zConcatDataset.cumsumrNcsZtt||_t|jdksJd|jD] }t|tr#Jdq||j|_dS)Nrz(datasets should not be an empty iterablez.ConcatDataset does not support IterableDataset) superr>rRrFrLrWrrarZ)r#rFd __class__r%r&r>s   zConcatDataset.__init__cCs |jdS)N)rZr5r%r%r&rD! zConcatDataset.__len__cCsf|dkr| t|krtdt||}t|j|}|dkr#|}n ||j|d}|j||S)Nrz8absolute value of index should not exceed dataset length)rLrPbisect bisect_rightrZrF)r#idxZ dataset_idxZ sample_idxr%r%r&r'$s zConcatDataset.__getitem__cCstjdtdd|jS)Nz:cummulative_sizes attribute is renamed to cumulative_sizes) stacklevel)warningswarnDeprecationWarningrZr5r%r%r&cummulative_sizes0szConcatDataset.cummulative_sizes)r1r2r3r4rrrrEint staticmethodrarr>rDr'propertyrq __classcell__r%r%rdr&rs    rcs>eZdZdZdeeddffdd ZddZd d ZZ S) ra_Dataset for chaining multiple :class:`IterableDataset` s. This class is useful to assemble different existing dataset streams. The chaining operation is done on-the-fly, so concatenating large-scale datasets with this class will be efficient. Args: datasets (iterable of IterableDataset): datasets to be chained together rFrNcst||_dSr-)rbr>rF)r#rFrdr%r&r>As  zChainDataset.__init__ccs.|jD]}t|tsJd|EdHqdS)N*ChainDataset only supports IterableDataset)rFrWr)r#rcr%r%r&r6Es   zChainDataset.__iter__cCs2d}|jD]}t|tsJd|t|7}q|S)Nrrv)rFrWrrL)r#totalrcr%r%r&rDJs  zChainDataset.__len__) r1r2r3r4rrr>r6rDrur%r%rdr&r7s  rc@sreZdZUdZeeed<eeed<deedeeddfddZ dd Z de ede efd d Z d d Z dS)rz Subset of a dataset at specified indices. Args: dataset (Dataset): The whole Dataset indices (sequence): Indices in the whole set selected for subset rNindicesrNcCs||_||_dSr-rNrx)r#rNrxr%r%r&r>]s zSubset.__init__cs2t|trjfdd|DSjj|S)Ncg|]}j|qSr%rx)r:ir5r%r& cz&Subset.__getitem__..)rWrRrNrx)r#rkr%r5r&r'as zSubset.__getitem__cs>ttjddrjfdd|DSfdd|DS)N __getitems__crzr%r{r:rkr5r%r&r}jr~z'Subset.__getitems__..csg|] }jj|qSr%ryrr5r%r&r}ls)callablegetattrrNr)r#rxr%r5r&rfszSubset.__getitems__cCs t|jSr-)rLrxr5r%r%r&rDnrgzSubset.__len__)r1r2r3r4rrrErrrr>r'rrrDr%r%r%r&rRs    rrNlengths generatorrc s&tt|drnt|dkrng}t|D]$\}}|dks |dkr(td|dttt|}||qtt|}t |D]}|t|}||d7<qE|}t|D]\}} | dkrmt d|dq\t|tkrztdt t||d fd d tt||DS) 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. rhrzFraction 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!)rcs&g|]\}}t|||qSr%)r)r:offsetlengthryr%r&r}s&z random_split..)mathisclosesum enumeraterPrrfloorrLr[rangernror tolistzipr) rNrrZsubset_lengthsr|fracZn_items_in_split remainderZ idx_to_add_atrr%ryr&rrs,   r)'rirnrtypingrrrrrrrr r r Ztorchr r Z torch._utilsrrr__all__rrstrZT_dictZT_tuplerrrrrrrrrrfloatrr%r%r%r&s40     "v/2!