VhB0(ddlZddlmZddlmZmZddlmZmZm Z m Z ddl Z ddl mZddlmZmZddlmZgdZe d d Zd efd ZedGddeeZedGddeZedGddeeZy)N) defaultdict)IteratorSized)AnyCallableOptionalTypeVar)functional_datapipe) DataChunk IterDataPipe)_check_unpickable_fn)BatcherIterDataPipeGrouperIterDataPipeUnBatcherIterDataPipe_T_coT) covariantnamec|dvretjd|d|dtdttj j jjj|Stdtd|) N)SHARDING_PRIORITIESShardingFilterIterDataPipe`zc` from `torch.utils.data.datapipes.iter.grouping` is going to be removed in PyTorch 2.1Please use `z5` from the `torch.utils.data.datapipes.iter.sharding`)category stacklevelzmodule z has no attribute ) warningswarn FutureWarninggetattrtorchutilsdata datapipesitershardingAttributeError__name__)rs X/home/dcms/DCMS/lib/python3.12/site-packages/torch/utils/data/datapipes/iter/grouping.py __getattr__r(s{ DD v& U W"  u{{''1166??FF 78*,>tfE FFbatchc eZdZUdZeed<eed<eed<defdededede eddf fd Z de efd Z defd Z xZS) ra2 Creates mini-batches of data (functional name: ``batch``). An outer dimension will be added as ``batch_size`` if ``drop_last`` is set to ``True``, or ``length % batch_size`` for the last batch if ``drop_last`` is set to ``False``. Args: datapipe: Iterable DataPipe being batched batch_size: The size of each batch drop_last: Option to drop the last batch if it's not full wrapper_class: wrapper to apply onto each batch (type ``List``) before yielding, defaults to ``DataChunk`` Example: >>> # xdoctest: +SKIP >>> from torchdata.datapipes.iter import IterableWrapper >>> dp = IterableWrapper(range(10)) >>> dp = dp.batch(batch_size=3, drop_last=True) >>> list(dp) [[0, 1, 2], [3, 4, 5], [6, 7, 8]] datapipe batch_size drop_lastF wrapper_classreturnNcr|dkDsJdt|||_||_||_||_y)Nrz+Batch size is required to be larger than 0!)super__init__r,r-r.r/)selfr,r-r.r/ __class__s r'r3zBatcherIterDataPipe.__init__As?A~LLL~   $"*r)c#Kg}|jD]A}|j|t||jk(s-|j |g}Ct|dkDr!|j s|j |yyywNr)r,appendlenr-r/r.)r4r*xs r'__iter__zBatcherIterDataPipe.__iter__Os~ A LLO5zT__,((//   u:>>>((//" s ;BABc8t|jtr`|jr"t |j|j zSt |j|j zdz |j zSt t|jd)Nz# instance doesn't have valid length) isinstancer,rr.r9r- TypeErrortyper&r4s r'__len__zBatcherIterDataPipe.__len__Zss dmmU +~~4==)T__<<DMM*T__>> # xdoctest: +SKIP >>> from torchdata.datapipes.iter import IterableWrapper >>> source_dp = IterableWrapper([[[0, 1], [2]], [[3, 4], [5]], [[6]]]) >>> dp1 = source_dp.unbatch() >>> list(dp1) [[0, 1], [2], [3, 4], [5], [6]] >>> dp2 = source_dp.unbatch(unbatch_level=2) >>> list(dp2) [0, 1, 2, 3, 4, 5, 6] r, unbatch_levelc ||_||_yN)r,rL)r4r,rLs r'r3zUnBatcherIterDataPipe.__init__|s  *r)c#|K|jD]'}|j||jEd{)y7w)NrL)r,_diverL)r4elements r'r;zUnBatcherIterDataPipe.__iter__s:}} MGzz'9K9KzL L L M Ls 0<:<c#vK|dkr td|dk(r>t|ttfr#|D]}|j |dEd{y|y|dk(r|yt|ttfr&|D] }|j ||dz Ed{"yt d|j d7k7"w)Nz unbatch_level must be -1 or >= 0rPrr=zunbatch_level z" exceeds the depth of the DataPipe) ValueErrorr>listr rQ IndexErrorrL)r4rRrLitems r'rQzUnBatcherIterDataPipe._dives 2 ?@ @ B 'D)#45#BD#zz$bzAAAB a M'D)#45#QD#zz$ma>OzPPPQ!$T%7%7$88Z[BQs%AB9 B5 A B9B7!B97B9N)r=) r&rCrDrEr rGr3r;rQr)r'rrds%,++c+Mr)rgroupbyceZdZdZdddddddeedeegefded e d e e d e e d efd Z dZ dZ ddZdZdZdZy)ra! Groups data from IterDataPipe by keys from ``group_key_fn``, yielding a ``DataChunk`` with batch size up to ``group_size``. (functional name: ``groupby``). The samples are read sequentially from the source ``datapipe``, and a batch of samples belonging to the same group will be yielded as soon as the size of the batch reaches ``group_size``. When the buffer is full, the DataPipe will yield the largest batch with the same key, provided that its size is larger than ``guaranteed_group_size``. If its size is smaller, it will be dropped if ``drop_remaining=True``. After iterating through the entirety of source ``datapipe``, everything not dropped due to the buffer capacity will be yielded from the buffer, even if the group sizes are smaller than ``guaranteed_group_size``. Args: datapipe: Iterable datapipe to be grouped group_key_fn: Function used to generate group key from the data of the source datapipe keep_key: Option to yield the matching key along with the items in a tuple, resulting in `(key, [items])` otherwise returning [items] buffer_size: The size of buffer for ungrouped data group_size: The max size of each group, a batch is yielded as soon as it reaches this size guaranteed_group_size: The guaranteed minimum group size to be yielded in case the buffer is full drop_remaining: Specifies if the group smaller than ``guaranteed_group_size`` will be dropped from buffer when the buffer is full Example: >>> import os >>> # xdoctest: +SKIP >>> from torchdata.datapipes.iter import IterableWrapper >>> def group_fn(file): ... return os.path.basename(file).split(".")[0] >>> source_dp = IterableWrapper(["a.png", "b.png", "a.json", "b.json", "a.jpg", "c.json"]) >>> dp0 = source_dp.groupby(group_key_fn=group_fn) >>> list(dp0) [['a.png', 'a.json', 'a.jpg'], ['b.png', 'b.json'], ['c.json']] >>> # A group is yielded as soon as its size equals to `group_size` >>> dp1 = source_dp.groupby(group_key_fn=group_fn, group_size=2) >>> list(dp1) [['a.png', 'a.json'], ['b.png', 'b.json'], ['a.jpg'], ['c.json']] >>> # Scenario where `buffer` is full, and group 'a' needs to be yielded since its size > `guaranteed_group_size` >>> dp2 = source_dp.groupby(group_key_fn=group_fn, buffer_size=3, group_size=3, guaranteed_group_size=2) >>> list(dp2) [['a.png', 'a.json'], ['b.png', 'b.json'], ['a.jpg'], ['c.json']] Fi'N)keep_key buffer_size group_sizeguaranteed_group_sizedrop_remainingr, group_key_fnr\r]r^r_r`c4t|||_||_||_||_t t |_d|_||_ d|_ ||d|cxkr|ksJJ||_ ||d|cxkr|ksJJ||_ ||_ t|_ yr7)r r,rar\max_buffer_sizerrVbuffer_elementscurr_buffer_sizer^r_r`r r/)r4r,rar\r]r^r_r`s r'r3zGrouperIterDataPipe.__init__s \*  (  *7B47H !$%)"  !k&=z0[0 00 00)3D & ,)a2G.U:.U UU.U UU)>D &,&r)cd}d}d}|jjD]8}t|j||kDst|j|}|}:|j=||jkr.|js"t dt |j||j||jk\r|j|}|xj|zc_|j|=|S)NrzFailed to group items)rdkeysr9r_r` RuntimeErrorstrre)r4 biggest_key biggest_sizeresult_to_yieldfindkeys r'_remove_biggest_keyz'GrouperIterDataPipe._remove_biggest_keys  ++002 &G4''01L@"4#7#7#@A %  &  & & 2t999'''T-A-A+-N)O   & & .t999"22;?O -   -r)c#K|jD]5}|j|}|j|j||xjdz c_|j |j t |j|k(rj|j|j|}|jr||fn||xjt |j|zc_|j|=|j|jk(s|j}||j|}|jr||fn|8t|jjD]^}|j|jj|}|xjt |zc_|jr||fn|`yw)Nr=)r,rardr8rer^r9r/r\rcrntuplergpop)r4r:keyresultrls r'r;zGrouperIterDataPipe.__iter__s EA##A&C   % , ,Q /  ! !Q & !*t#$$S)C0*.););Drs#+33/EGH  4( Gc GW;Y,y1;Y;Y|Y1L1 1hYh%,y1h% h%r)