x-j6dddlmZddlmZmZmZmZddlZddl m Z ddl m Z erddl mZmZmZmZddlmZddlmZed ZGd d eeZGd d eeZGddeeZddZGddeeZGddeeZdS)) annotations) TYPE_CHECKINGAnyGenericTypeVarN)core)randperm) GeneratorIteratorSequenceSized)Tensor_Tc@eZdZUdZded<d d dZdd Zerdd ZdSdS)Samplera An abstract class to encapsulate methods and behaviors of samplers. All sampler used by :code:`paddle.io.BatchSampler` should be a subclass of :code:`paddle.io.Sampler`, BatchSampler subclasses should implement following methods: :code:`__iter__`: return sample index iterably, which iterate over indices of dataset elements :code:`__len__`: the number of sample in :attr:`data_source` Args: data_source(Dataset, optional): this could be an instance of :code:`paddle.io.Dataset` other Python object which implemented :code:`__len__` for Sampler to get indices as the range of :attr:`dataset` length. Default None. Returns: Sampler: an iterable object for sample indices iterating Examples: .. code-block:: python >>> import numpy as np >>> from paddle.io import Dataset, Sampler >>> class RandomDataset(Dataset): # type: ignore[type-arg] ... def __init__(self, num_samples): ... self.num_samples = num_samples ... ... def __getitem__(self, idx): ... image = np.random.random([784]).astype('float32') ... label = np.random.randint(0, 9, (1, )).astype('int64') ... return image, label ... ... def __len__(self): ... return self.num_samples ... >>> class MySampler(Sampler): # type: ignore[type-arg] ... def __init__(self, data_source): ... self.data_source = data_source ... ... def __iter__(self): ... return iter(range(len(self.data_source))) # type: ignore[arg-type] ... ... def __len__(self): ... return len(self.data_source) # type: ignore[arg-type] ... >>> sampler = MySampler(data_source=RandomDataset(100)) >>> for index in sampler: ... print(index) 0 1 2 ... 99 see `paddle.io.BatchSampler` see `paddle.io.DataLoader` Sized | None data_sourceNreturnNonec||_dSNrselfrs \/var/www/html/banglarbhumi/venv/lib/python3.11/site-packages/paddle/io/dataloader/sampler.py__init__zSampler.__init__m& Iterator[_T]ctr)NotImplementedErrorrs r__iter__zSampler.__iter__ps!!rintcdSrr#s r__len__zSampler.__len__wsrr)rrrr)rr rr%) __name__ __module__ __qualname____doc____annotations__rr$rr(r'rrrr(sz@@D'''''"""" &%%%%%%&&rrc6eZdZUdZded<d dZd dZdd Zd S)SequenceSamplera4 Iterate samples sequentially, yield :code:`0, 1, 2, ..., len(data_source) -1` generally, Args: data_source(Dataset): dataset to sample, this could be an instance of :code:`paddle.io.Dataset` other Python object which implemented :code:`__len__`. Returns: Sampler: a Sampler yield sample index sequentially Examples: .. code-block:: python >>> import numpy as np >>> from paddle.io import Dataset, SequenceSampler >>> class RandomDataset(Dataset): # type: ignore[type-arg] ... def __init__(self, num_samples): ... self.num_samples = num_samples ... ... def __getitem__(self, idx): ... image = np.random.random([784]).astype('float32') ... label = np.random.randint(0, 9, (1, )).astype('int64') ... return image, label ... ... def __len__(self): ... return self.num_samples ... >>> sampler = SequenceSampler(data_source=RandomDataset(100)) >>> for index in sampler: ... print(index) 0 1 2 ... 99 see `paddle.io.Sampler` rrrrc||_dSrrrs rrzSequenceSampler.__init__rr Iterator[int]c^ttt|jSr)iterrangelenrr#s rr$zSequenceSampler.__iter__s#E#d.//00111rr%c*t|jSr)r6rr#s rr(zSequenceSampler.__len__s4#$$$rN)rrrrrr2r)r*r+r,r-r.rr$r(r'rrr0r0zsk**X''''2222%%%%%%rr0cjeZdZUdZded<ded<ded< dddZeddZddZddZ d S) RandomSamplera Iterate samples randomly, yield shuffled indices, if :attr:`replacement=False`, yield shuffled indices of the whole data source, if :attr:`replacement=True`, :attr:`num_samples` can set to specify the sample number to draw. Args: data_source(Dataset): dataset to sample, this could be an instance of :ref:`api_paddle_io_Dataset` or :ref:`api_paddle_io_IterableDataset` or other Python object which implemented :code:`__len__` to get indices as the range of :code:`dataset` length. Default None. replacement(bool, optional): If False, sample the whole dataset, If True, set :attr:`num_samples` for how many samples to draw. Default False. num_samples(int, optional): set sample number to draw. Default None, which is set to the length of `data_source`. generator(Generator, optional): specify a generator to sample the :code:`data_source`. Default None, disabled. Returns: RandomSampler: a Sampler yield sample index randomly. Examples: .. code-block:: python >>> import numpy as np >>> from paddle.io import Dataset, RandomSampler >>> np.random.seed(2023) >>> class RandomDataset(Dataset): # type: ignore[type-arg] ... def __init__(self, num_samples): ... self.num_samples = num_samples ... ... def __getitem__(self, idx): ... image = np.random.random([784]).astype('float32') ... label = np.random.randint(0, 9, (1, )).astype('int64') ... return image, label ... ... def __len__(self): ... return self.num_samples ... >>> sampler = RandomSampler(data_source=RandomDataset(100)) >>> for index in sampler: ... print(index) 56 12 68 ... 87 rrbool replacement!Generator[int, None, None] | None generatorFN num_samples int | Nonerrc||_||_||_||_t |jt st d|j|jsI|jt|jkr,td|jdt|jt |jtr |jdkrtd|jdS)Nz:expect boolean value for replacement, but got replacement=zunum_samples should be smaller than or equal to length of data_source when replacement is False, but got num_samples: z > data_source: rz>num_samples should be a positive integer, but got num_samples=) rr= _num_samplesr? isinstancer< TypeErrorr@r6 ValueErrorr%)rrr=r@r?s rrzRandomSampler.__init__s'&'"$*D11 2#/22   D$4s4;K7L7L$L$Lb(,(8bbJMdN^J_J_bb  $*C00 D4D4I4I:'+'7:: 5J4Irr%cF|jt|jS|jSr)rCr6rr#s rr@zRandomSampler.num_sampless%   $t'(( (  rr2c#$Kt|j}|jrEt|jD].} t |j}n#t $rYdSwxYw|V/dS|jrTtj tj ||jd D]}|VdStj tj ||jd D]}|VdS)NT)replaceF) r6rr?r5r@next StopIterationr=nprandomchoicearangetolist)rniindexs rr$zRandomSampler.__iter__ sH  ! ! > 4+,,   00EE$FFF    Y--IaLL$"2D.&((  E KKKK   Y--IaLL$"2E.&((  E KKKK  sA  AAc|jSr)r@r#s rr(zRandomSampler.__len__!s r)FNN) rrr=r<r@rAr?r>rrr)r8) r*r+r,r-r.rpropertyr@r$r(r'rrr;r;s..`0000 ""&7; <!!!X!     *      rr;Tcjt|tjr|}t|tt frt j|}t|t js Jdt|j dks Jd| d|j df}t j |dks Jdt j |t jkr Jdt j |t jkr Jdt j|dkd }t j |d ks Jd |s"t j ||ks Jd ||d z }g}t#|j d D]J}t j|j d ||||}||Kt j|S)Nz=weights should be paddle.Tensor, numpy.ndarray, list or tuplez$weights should be a 1-D or 2-D arraygz weights should be positive valuezweights should not be INFzweights should not be NaN)axisrz#weights should have positive valueszUweights positive value number should not less than num_samples when replacement=False)rDr DenseTensornumpylisttuplerLarrayndarrayr6shapereshapeallanyinfnansumr5rMrNappend)weightsr@r= non_zerosretsrRrets r_weighted_samplerm%s"'4+,,"--//'D%=))$(7## grz * *G * w}   " " "$J " " "oor7=#4566G 6'S. ! !EE#EEE !vg'((EE*EEE (vg'((EE*EEE (w}1---I 6)a- GG"GGG  vi;.//   ;  /   +++G D 7=# $ $i M! k;    C 8D>>rcNeZdZUdZded<ded<ded< ddd Zdd ZddZdS)WeightedRandomSampleraH Random sample with given weights (probabilities), sample index will be in range [0, len(weights) - 1], if :attr:`replacement` is True, index can be sampled multiple times. Args: weights(numpy.ndarray|paddle.Tensor|list|tuple): sequence of weights, should be numpy array, paddle.Tensor, list or tuple num_samples(int): set sample number to draw from sampler. replacement(bool): Whether to draw sample with replacements, default True Returns: Sampler: a Sampler yield sample index randomly by given weights Examples: .. code-block:: python >>> import numpy as np >>> from paddle.io import WeightedRandomSampler >>> np.random.seed(2023) >>> sampler = WeightedRandomSampler( ... weights=[0.1, 0.3, 0.5, 0.7, 0.2], ... num_samples=5, ... replacement=True ... ) >>> for index in sampler: ... print(index) 2 4 3 1 1 +npt.NDArray[Any] | Tensor | Sequence[float]rir%r@r<r=Trrct|tr|dkrtdt|tstd||_||_||_dS)Nrz(num_samples should be a positive integerz%replacement should be a boolean value)rDr%rFr<rir@r=)rrir@r=s rrzWeightedRandomSampler.__init__nso +s++ I{a/?/?GHH H+t,, FDEE E &&rr2ct|j|j|j}t |dSNrX)rmrir@r=r4rbrP)ridxss rr$zWeightedRandomSampler.__iter__|sH L$*D,<  DLL$$++--...rcxtj|jj|jjdz}|j|zSrs)rLprodrirar@)rmuls rr(zWeightedRandomSampler.__len__s3gdl())T\-?-CC#%%rNT)rirpr@r%r=r<rrr8r)r9r'rrroroEs""H9888 ! ' ' ' ' '//// &&&&&&rroc6eZdZUdZded<d dZd dZdd Zd S)SubsetRandomSamplera Randomly sample elements from a given list of indices, without replacement. Args: indices (sequence): a sequence of indices Examples: .. code-block:: python >>> import paddle >>> from paddle.io import SubsetRandomSampler >>> paddle.seed(2023) >>> sampler = SubsetRandomSampler(indices=[1, 3, 5, 7, 9]) >>> for index in sampler: ... print(index) 9 3 7 5 1 Sequence[int]indicesrrcXt|dkrtd||_dS)NrzHThe length of `indices` in SubsetRandomSampler should be greater than 0.)r6rFr|)rr|s rrzSubsetRandomSampler.__init__s5 w<<1  Z  rr2c#pKtt|jD]}|j|VdSr)r r6r|)rrRs rr$zSubsetRandomSampler.__iter__sF#dl++,, " "A,q/ ! ! ! ! " "rr%c*t|jSr)r6r|r#s rr(zSubsetRandomSampler.__len__s4<   rN)r|r{rrr8r)r9r'rrrzrzsj4""""!!!!!!rrzrx) __future__rtypingrrrrr\rL frameworkr tensorr collections.abcr r r r numpy.typingnptpaddlerrrr%r0r;rmrorzr'rrrs#"""""DDDDDDDDDDDD WT]]O&O&O&O&O&gbkO&O&O&d6%6%6%6%6%gcl6%6%6%ro o o o o GCLo o o d@?&?&?&?&?&GCL?&?&?&D)!)!)!)!)!'#,)!)!)!)!)!r