§
    ·¯¸j¤  ã                   óf   — d dl Z d dlZd dlmZ d dlmZ  G d„ de¦  «        Z G d„ de¦  «        ZdS )é    N)ÚSamplerc                   ó&   — e Zd ZdZdd„Zd„ Zd„ ZdS )ÚOrderedDistributedSamplera€  Sampler that restricts data loading to a subset of the dataset.
    It is especially useful in conjunction with
    :class:`torch.nn.parallel.DistributedDataParallel`. In such case, each
    process can pass a DistributedSampler instance as a DataLoader sampler,
    and load a subset of the original dataset that is exclusive to it.
    .. note::
        Dataset is assumed to be of constant size.
    Arguments:
        dataset: Dataset used for sampling.
        num_replicas (optional): Number of processes participating in
            distributed training.
        rank (optional): Rank of the current process within num_replicas.
    Nc                 óº  — |€5t          j        ¦   «         st          d¦  «        ‚t          j        ¦   «         }|€5t          j        ¦   «         st          d¦  «        ‚t          j        ¦   «         }|| _        || _        || _        t          t          j
        t          | j        ¦  «        dz  | j        z  ¦  «        ¦  «        | _        | j        | j        z  | _        d S )Nú,Requires distributed package to be availableg      ð?)ÚdistÚis_availableÚRuntimeErrorÚget_world_sizeÚget_rankÚdatasetÚnum_replicasÚrankÚintÚmathÚceilÚlenÚnum_samplesÚ
total_size)Úselfr   r   r   s       úm/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/timm/data/distributed_sampler.pyÚ__init__z"OrderedDistributedSampler.__init__   sÅ   € ØÐÝÔ$Ñ&Ô&ð SÝ"Ð#QÑRÔRÐRÝÔ.Ñ0Ô0ˆLØˆ<ÝÔ$Ñ&Ô&ð SÝ"Ð#QÑRÔRÐRÝ”=‘?”?ˆDØˆŒØ(ˆÔØˆŒ	Ý�tœy­¨T¬\Ñ):Ô):¸SÑ)@À4ÔCTÑ)TÑUÔUÑVÔVˆÔØÔ*¨TÔ->Ñ>ˆŒˆˆó    c                 ó\  — t          t          t          | j        ¦  «        ¦  «        ¦  «        }||d | j        t          |¦  «        z
  …         z  }t          |¦  «        | j        k    sJ ‚|| j        | j        | j        …         }t          |¦  «        | j        k    sJ ‚t          |¦  «        S ©N)	ÚlistÚranger   r   r   r   r   r   Úiter)r   Úindicess     r   Ú__iter__z"OrderedDistributedSampler.__iter__%   sž   € Ý•u�S ¤Ñ.Ô.Ñ/Ô/Ñ0Ô0ˆð 	�7Ð<˜Tœ_­s°7©|¬|Ñ;Ð<Ô=Ñ=ˆÝ�7‰|Œ|˜tœÒ.Ð.Ð.Ð.ð ˜$œ) D¤O°DÔ4EÐEÔFˆÝ�7‰|Œ|˜tÔ/Ò/Ð/Ð/Ð/å�G‰}Œ}Ðr   c                 ó   — | j         S r   )r   ©r   s    r   Ú__len__z!OrderedDistributedSampler.__len__2   s   € ØÔÐr   )NN)Ú__name__Ú
__module__Ú__qualname__Ú__doc__r   r    r#   © r   r   r   r      sP   € € € € € ðð ð?ð ?ð ?ð ?ðð ð ð ð  ð  ð  ð  r   r   c                   ó8   — e Zd ZdZ	 	 	 	 	 	 dd„Zd„ Zd	„ Zd
„ ZdS )ÚRepeatAugSamplera»  Sampler that restricts data loading to a subset of the dataset for distributed,
    with repeated augmentation.
    It ensures that different each augmented version of a sample will be visible to a
    different process (GPU). Heavily based on torch.utils.data.DistributedSampler

    This sampler was taken from https://github.com/facebookresearch/deit/blob/0c4b8f60/samplers.py
    Used in
    Copyright (c) 2015-present, Facebook, Inc.
    NTé   é   r   c                 óì  — |€5t          j        ¦   «         st          d¦  «        ‚t          j        ¦   «         }|€5t          j        ¦   «         st          d¦  «        ‚t          j        ¦   «         }|| _        || _        || _        || _        || _	        d| _
        t          t          j        t          | j        ¦  «        |z  | j        z  ¦  «        ¦  «        | _        | j        | j        z  | _        |p|}|rCt          t          j        t          | j        ¦  «        |z  |z  |z  ¦  «        ¦  «        | _        d S t          t          j        t          | j        ¦  «        |z  ¦  «        ¦  «        | _        d S )Nr   r   )r   r	   r
   r   r   r   r   r   ÚshuffleÚnum_repeatsÚepochr   r   r   r   r   r   ÚfloorÚnum_selected_samples)r   r   r   r   r.   r/   Úselected_roundÚselected_ratios           r   r   zRepeatAugSampler.__init__A   se  € ð ÐÝÔ$Ñ&Ô&ð SÝ"Ð#QÑRÔRÐRÝÔ.Ñ0Ô0ˆLØˆ<ÝÔ$Ñ&Ô&ð SÝ"Ð#QÑRÔRÐRÝ”=‘?”?ˆDØˆŒØ(ˆÔØˆŒ	ØˆŒØ&ˆÔØˆŒ
Ý�tœy­¨T¬\Ñ):Ô):¸[Ñ)HÈ4ÔK\Ñ)\Ñ]Ô]Ñ^Ô^ˆÔØÔ*¨TÔ->Ñ>ˆŒð (Ð7¨<ˆØð 	[Ý(+­D¬JÝ�T”\Ñ"Ô" nÑ4°~ÑEÈÑVñ-Xô -Xñ )Yô )YˆDÔ%Ð%Ð%õ ),­D¬Iµc¸$¼,Ñ6GÔ6GÈ.Ñ6XÑ,YÔ,YÑ(ZÔ(ZˆDÔ%Ð%Ð%r   c                 óÚ  ‡ — t          j        ¦   «         }|                     ‰ j        ¦  «         ‰ j        r)t          j        t          ‰ j        ¦  «        |¬¦  «        }n(t          j        dt          ‰ j        ¦  «        ¬¦  «        }t          ‰ j
        t          ¦  «        r{‰ j
                             ¦   «         sbt          j        ‰ j
        t          ‰ j        ¦  «        z  ¦  «        }|t          j        ˆ fd„t!          |¦  «        D ¦   «         ¦  «                 }n)t          j        |t%          ‰ j
        ¦  «        d¬¦  «        }|                     ¦   «         }‰ j        t          |¦  «        z
  }|dk    r||d |…         z  }t          |¦  «        ‰ j        k    sJ ‚|‰ j        ‰ j        ‰ j        …         }t          |¦  «        ‰ j        k    sJ ‚t1          |d ‰ j        …         ¦  «        S )N)Ú	generatorr   )ÚstartÚendc                 ó>   •— g | ]}t          |‰j        z  ¦  «        ‘ŒS r(   )r   r/   )Ú.0Úir   s     €r   ú
<listcomp>z-RepeatAugSampler.__iter__.<locals>.<listcomp>r   s)   ø€ Ð+cÐ+cÐ+cÈ1­C°°TÔ5EÑ0EÑ,FÔ,FÐ+cÐ+cÐ+cr   )ÚrepeatsÚdim)ÚtorchÚ	GeneratorÚmanual_seedr0   r.   Úrandpermr   r   ÚarangeÚ
isinstancer/   ÚfloatÚ
is_integerr   r   Útensorr   Úrepeat_interleaver   Útolistr   r   r   r   r   r2   )r   Úgr   Úrepeat_sizeÚpadding_sizes   `    r   r    zRepeatAugSampler.__iter__e   s¿  ø€ åŒOÑÔˆØ	�Š�d”jÑ!Ô!Ð!ØŒ<ð 	CÝ”n¥S¨¬Ñ%6Ô%6À!ÐDÑDÔDˆGˆGå”l¨µ°D´LÑ0AÔ0AÐBÑBÔBˆGõ �dÔ&­Ñ.Ô.ð 	]°tÔ7G×7RÒ7RÑ7TÔ7Tð 	]åœ) DÔ$4µs¸4¼<Ñ7HÔ7HÑ$HÑIÔIˆKØ�eœlÐ+cÐ+cÐ+cÐ+cÕPUÐVaÑPbÔPbÐ+cÑ+cÔ+cÑdÔdÔeˆGˆGåÔ-¨g½sÀ4ÔCSÑ?TÔ?TÐZ[Ð\Ñ\Ô\ˆGØ—.’.Ñ"Ô"ˆà”­¨W©¬Ñ5ˆØ˜!ÒÐØ�w˜} ˜}Ô-Ñ-ˆGÝ�7‰|Œ|˜tœÒ.Ð.Ð.Ð.ð ˜$œ) D¤O°DÔ4EÐEÔFˆÝ�7‰|Œ|˜tÔ/Ò/Ð/Ð/Ð/õ �GÐ6˜TÔ6Ð6Ô7Ñ8Ô8Ð8r   c                 ó   — | j         S r   )r2   r"   s    r   r#   zRepeatAugSampler.__len__ƒ   s   € ØÔ(Ð(r   c                 ó   — || _         d S r   )r0   )r   r0   s     r   Ú	set_epochzRepeatAugSampler.set_epoch†   s   € ØˆŒ
ˆ
ˆ
r   )NNTr+   r,   r   )r$   r%   r&   r'   r   r    r#   rO   r(   r   r   r*   r*   6   sx   € € € € € ðð ð ØØØØØð"[ð "[ð "[ð "[ðH9ð 9ð 9ð<)ð )ð )ðð ð ð ð r   r*   )	r   r?   Útorch.utils.datar   Útorch.distributedÚdistributedr   r   r*   r(   r   r   ú<module>rS      s¤   ðØ €€€Ø €€€Ø $Ð $Ð $Ð $Ð $Ð $Ø  Ð  Ð  Ð  Ð  Ð  ð, ð , ð , ð , ð ,  ñ , ô , ð , ð^Qð Qð Qð Qð Q�wñ Qô Qð Qð Qð Qr   