tjԠddlmZddlZddlmZddlmZmZddlm Z m Z m Z ddl m ZddlmZdd lmZgd ZGd d eZGd deZGddeeZGddeZGddeeZGddeZGddeeZGddeZGddeeZGddeZ dS))AnyN)Tensor) functionalinit) ParameterUninitializedBufferUninitializedParameter) SyncBatchNorm)LazyModuleMixin)Module) BatchNorm1dLazyBatchNorm1d BatchNorm2dLazyBatchNorm2d BatchNorm3dLazyBatchNorm3dr ceZdZUdZdZgdZeed<eed<edzed<e ed<e ed < dd d dedededzde d e de ddffdZ ddZ ddZ dZ dZ dfd ZxZS) _NormBasez,Common base of _InstanceNorm and _BatchNorm.)track_running_statsmomentumeps num_featuresaffinerrNrrrh㈵>皙?Tbiasrreturnc ||d} t||_||_||_||_||_|jrbttj |fi| |_ |r%ttj |fi| |_ nC| ddn,| dd| dd|jr| dtj|fi| | dtj|fi| ||| dtj d dtjid | D|nB| dd| dd| dd|dS) Ndevicedtyperweight running_mean running_varnum_batches_trackedrr$c&i|]\}}|dk ||Sr$.0kvs X/var/www/html/CA-Chatbot/venv/lib/python3.11/site-packages/torch/nn/modules/batchnorm.py z&_NormBase.__init__..Qs#OOO1!w,,q!,,,r)super__init__rrrrrrtorchemptyr%rregister_parameterregister_bufferzerosonestensorlongitemsreset_parameters selfrrrrrr#r$rfactory_kwargs __class__s r0r5z_NormBase.__init__&s(%+U;; (   #6 ; 2#EK $O$O$O$OPPDK 6%ek,&Q&Q.&Q&QRR ''5555  # #Hd 3 3 3  # #FD 1 1 1  # >   L K KN K K     uz,II.II     ,  +  % *PO(<(<(>(>OOO      3 3   6 6 6   5 5 5  !6 = = = r2c|jrN|j|jd|jdSdS)Nr )rr&zero_r'fill_r(rAs r0reset_running_statsz_NormBase.reset_running_stats[s`  # -   # # % % %   " "1 % % %  $ * * , , , , ,  - -r2c||jr;tj|j|jtj|jdSdSdSN)rHrrones_r%rzeros_rGs r0r?z_NormBase.reset_parameterscs`   """ ; ' Jt{ # # #y$ DI&&&&& ' '$$r2ctrJ)NotImplementedErrorrAinputs r0_check_input_dimz_NormBase._check_input_dimjs!!r2c<djdi|jd|jduiS)Nz{{num_features}, eps={eps}, momentum={momentum}, affine={affine}, bias={use_bias}, track_running_stats={track_running_stats}use_biasr+)format__dict__rrGs r0 extra_reprz_NormBase.extra_reprmsH P IIO  -  *.)4*?    r2c X|dd}||dkrc|jr\|dz} | |vrS|j)|jjt jdkr|jnt jdtj|| <t|||||||dS)Nversionrr(metarr*) getrr(r#r6r<r=r4_load_from_state_dict) rA state_dictprefixlocal_metadatastrict missing_keysunexpected_keys error_msgsrXnum_batches_tracked_keyrCs r0r[z_NormBase._load_from_state_dictus!$$Y55 Ow{{0H{'-/D&D #&j88/;075<;O;OOO,,auz::: 23 %%            r2rrTTNNr N)__name__ __module__ __qualname____doc___version __constants__int__annotations__floatboolr5rHr?rQrVr[ __classcell__rCs@r0rrsn66HXXXM JJJdl LLL!$$(3 3 3 3 3 3 $, 3  3 " 3 3  3 3 3 3 3 3 j----''''"""                 r2rcfeZdZ ddddedededzd ed ed ed dffd Zded efdZxZ S) _BatchNormrrTNrrrrrrrr cV||d} tj|||||fi| d|idSNr"r)r4r5r@s r0r5z_BatchNorm.__init__si%+U;;                 r2rPc |||jd}n|j}|jrN|jrG|j@|jd|jdt |jz }n|j} |jrd}n|jduo|jdu} tj ||jr|jr|jnd|jr|jr|jnd|j |j |||j S)Nr ?T)rQrtrainingrr(add_rnr&r'F batch_normr%rr)rArPexponential_average_factor bn_trainings r0forwardz_BatchNorm.forwards1 e$$$ = ), & &)- & = ?T5 ?'3(--a000=(14uT=U7V7V1V..15.  = UKK,4T4;Kt;SK | }(,(@!!$(M WT5M WD  SW K I  & H   r2rd) rfrgrhrlrnror5rrrprqs@r0rsrss!$$(      $,    "          .0 V0 0 0 0 0 0 0 0 0 r2rsc`eZdZUeed<eed< d dd d fd Zd fd Zd d ZxZS) _LazyNormBaser%rrrTNrr c||d}tjd||ddfi|ddi||_||_|jr$t di||_|rt di||_|jrctdi||_tdi||_ tj ddtj id| D|_dSdS) Nr"rFrr$c&i|]\}}|dk ||Sr*r+r,s r0r1z*_LazyNormBase.__init__..s#KKKDAqa7ll1alllr2r+r3)r4r5rrr r%rrr&r'r6r<r=r>r() rArrrrr#r$rrBrCs r0r5z_LazyNormBase.__init__sB%+U;;         #6 ; E0BB>BBDK E2DD^DD  #  3 E En E ED 2DD^DDD ',|((j(LKN$8$8$:$:KKK ((D $ $ $  r2c|s-|jdkr$tdSdSdS)Nr)has_uninitialized_paramsrr4r?)rArCs r0r?z_LazyNormBase.reset_parameterssP,,.. '43D3I3I GG $ $ & & & & & ' '3I3Ir2cN|r|jd|_|jrt |jt std|j|jf|j It |j t std|j |jf|j r@|j |jf|j |jf| dSdS)Nr z-self.weight must be an UninitializedParameterz+self.bias must be an UninitializedParameter)rshaperr isinstancer%r AssertionError materializerrr&r'r?rOs r0initialize_parametersz#_LazyNormBase.initialize_parameterss@  ( ( * * $ % AD { @!$+/EFF(G ''):(<===9(%di1GHH,II))4+<*>???' !--&( ,,&(  ! ! # # # # #+ $ $r2rdre) rfrgrhr rmr5r?rrprqs@r0rrs """"      *** ******X'''''' $$$$$$$$r2rceZdZdZddZdS)raApplies Batch Normalization over a 2D or 3D input. Method described in the paper `Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift `__ . .. math:: y = \frac{x - \mathrm{E}[x]}{\sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta The mean and standard-deviation are calculated per-dimension over the mini-batches and :math:`\gamma` and :math:`\beta` are learnable parameter vectors of size `C` (where `C` is the number of features or channels of the input). By default, the elements of :math:`\gamma` are set to 1 and the elements of :math:`\beta` are set to 0. At train time in the forward pass, the variance is calculated via the biased estimator, equivalent to ``torch.var(input, correction=0)``. However, the value stored in the moving average of the variance is calculated via the unbiased estimator, equivalent to ``torch.var(input, correction=1)``. Also by default, during training this layer keeps running estimates of its computed mean and variance, which are then used for normalization during evaluation. The running estimates are kept with a default :attr:`momentum` of 0.1. If :attr:`track_running_stats` is set to ``False``, this layer then does not keep running estimates, and batch statistics are instead used during evaluation time as well. .. note:: This :attr:`momentum` argument is different from one used in optimizer classes and the conventional notion of momentum. Mathematically, the update rule for running statistics here is :math:`\hat{x}_\text{new} = (1 - \text{momentum}) \times \hat{x} + \text{momentum} \times x_t`, where :math:`\hat{x}` is the estimated statistic and :math:`x_t` is the new observed value. Because the Batch Normalization is done over the `C` dimension, computing statistics on `(N, L)` slices, it's common terminology to call this Temporal Batch Normalization. Args: num_features: number of features or channels :math:`C` of the input eps: a value added to the denominator for numerical stability. Default: 1e-5 momentum: the value used for the running_mean and running_var computation. Can be set to ``None`` for cumulative moving average (i.e. simple average). Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters. Default: ``True`` track_running_stats: a boolean value that when set to ``True``, this module tracks the running mean and variance, and when set to ``False``, this module does not track such statistics, and initializes statistics buffers :attr:`running_mean` and :attr:`running_var` as ``None``. When these buffers are ``None``, this module always uses batch statistics. in both training and eval modes. Default: ``True`` bias: If set to ``False``, the layer will not learn an additive bias (only relevant if :attr:`affine` is ``True``). Default: ``True`` Shape: - Input: :math:`(N, C)` or :math:`(N, C, L)`, where :math:`N` is the batch size, :math:`C` is the number of features or channels, and :math:`L` is the sequence length - Output: :math:`(N, C)` or :math:`(N, C, L)` (same shape as input) Examples:: >>> # With Learnable Parameters >>> m = nn.BatchNorm1d(100) >>> # Without Learnable Parameters >>> m = nn.BatchNorm1d(100, affine=False) >>> input = torch.randn(20, 100) >>> output = m(input) r Nc|dkr=|dkr'td|ddSdSNrzexpected 2D or 3D input (got D input)dim ValueErrorrOs r0rQzBatchNorm1d._check_input_dim{W 99;;!   q 0 0RUYY[[RRRSS S   0 0r2rerfrgrhrirQr+r2r0rr2s;FFPTTTTTTr2rceZdZdZeZddZdS)raA :class:`torch.nn.BatchNorm1d` module with lazy initialization. Lazy initialization based on the ``num_features`` argument of the :class:`BatchNorm1d` that is inferred from the ``input.size(1)``. The attributes that will be lazily initialized are `weight`, `bias`, `running_mean` and `running_var`. Check the :class:`torch.nn.modules.lazy.LazyModuleMixin` for further documentation on lazy modules and their limitations. Args: eps: a value added to the denominator for numerical stability. Default: 1e-5 momentum: the value used for the running_mean and running_var computation. Can be set to ``None`` for cumulative moving average (i.e. simple average). Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters. Default: ``True`` track_running_stats: a boolean value that when set to ``True``, this module tracks the running mean and variance, and when set to ``False``, this module does not track such statistics, and initializes statistics buffers :attr:`running_mean` and :attr:`running_var` as ``None``. When these buffers are ``None``, this module always uses batch statistics. in both training and eval modes. Default: ``True`` bias: If set to ``False``, the layer will not learn an additive bias (only relevant if :attr:`affine` is ``True``). Default: ``True`` r Nc|dkr=|dkr'td|ddSdSrrrOs r0rQz LazyBatchNorm1d._check_input_dimrr2re)rfrgrhrir cls_to_becomerQr+r2r0rrs?8 MTTTTTTr2rceZdZdZddZdS)ra2Applies Batch Normalization over a 4D input. 4D is a mini-batch of 2D inputs with additional channel dimension. Method described in the paper `Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift `__ . .. math:: y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta The mean and standard-deviation are calculated per-dimension over the mini-batches and :math:`\gamma` and :math:`\beta` are learnable parameter vectors of size `C` (where `C` is the input size). By default, the elements of :math:`\gamma` are set to 1 and the elements of :math:`\beta` are set to 0. At train time in the forward pass, the standard-deviation is calculated via the biased estimator, equivalent to ``torch.var(input, correction=0)``. However, the value stored in the moving average of the standard-deviation is calculated via the unbiased estimator, equivalent to ``torch.var(input, correction=1)``. Also by default, during training this layer keeps running estimates of its computed mean and variance, which are then used for normalization during evaluation. The running estimates are kept with a default :attr:`momentum` of 0.1. If :attr:`track_running_stats` is set to ``False``, this layer then does not keep running estimates, and batch statistics are instead used during evaluation time as well. .. note:: This :attr:`momentum` argument is different from one used in optimizer classes and the conventional notion of momentum. Mathematically, the update rule for running statistics here is :math:`\hat{x}_\text{new} = (1 - \text{momentum}) \times \hat{x} + \text{momentum} \times x_t`, where :math:`\hat{x}` is the estimated statistic and :math:`x_t` is the new observed value. Because the Batch Normalization is done over the `C` dimension, computing statistics on `(N, H, W)` slices, it's common terminology to call this Spatial Batch Normalization. Args: num_features: :math:`C` from an expected input of size :math:`(N, C, H, W)` eps: a value added to the denominator for numerical stability. Default: 1e-5 momentum: the value used for the running_mean and running_var computation. Can be set to ``None`` for cumulative moving average (i.e. simple average). Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters. Default: ``True`` track_running_stats: a boolean value that when set to ``True``, this module tracks the running mean and variance, and when set to ``False``, this module does not track such statistics, and initializes statistics buffers :attr:`running_mean` and :attr:`running_var` as ``None``. When these buffers are ``None``, this module always uses batch statistics. in both training and eval modes. Default: ``True`` bias: If set to ``False``, the layer will not learn an additive bias (only relevant if :attr:`affine` is ``True``). Default: ``True`` Shape: - Input: :math:`(N, C, H, W)` - Output: :math:`(N, C, H, W)` (same shape as input) Examples:: >>> # With Learnable Parameters >>> m = nn.BatchNorm2d(100) >>> # Without Learnable Parameters >>> m = nn.BatchNorm2d(100, affine=False) >>> input = torch.randn(20, 100, 35, 45) >>> output = m(input) r Nc|dkr%td|ddSNzexpected 4D input (got rrrOs r0rQzBatchNorm2d._check_input_dim? 99;;!  Luyy{{LLLMM M  r2rerr+r2r0rr;GGRNNNNNNr2rceZdZdZeZddZdS)raA :class:`torch.nn.BatchNorm2d` module with lazy initialization. Lazy initialization is done for the ``num_features`` argument of the :class:`BatchNorm2d` that is inferred from the ``input.size(1)``. The attributes that will be lazily initialized are `weight`, `bias`, `running_mean` and `running_var`. Check the :class:`torch.nn.modules.lazy.LazyModuleMixin` for further documentation on lazy modules and their limitations. Args: eps: a value added to the denominator for numerical stability. Default: 1e-5 momentum: the value used for the running_mean and running_var computation. Can be set to ``None`` for cumulative moving average (i.e. simple average). Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters. Default: ``True`` track_running_stats: a boolean value that when set to ``True``, this module tracks the running mean and variance, and when set to ``False``, this module does not track such statistics, and initializes statistics buffers :attr:`running_mean` and :attr:`running_var` as ``None``. When these buffers are ``None``, this module always uses batch statistics. in both training and eval modes. Default: ``True`` bias: If set to ``False``, the layer will not learn an additive bias (only relevant if :attr:`affine` is ``True``). Default: ``True`` r Nc|dkr%td|ddSrrrOs r0rQz LazyBatchNorm2d._check_input_dimrr2re)rfrgrhrirrrQr+r2r0rr?8 MNNNNNNr2rceZdZdZddZdS)ragApplies Batch Normalization over a 5D input. 5D is a mini-batch of 3D inputs with additional channel dimension as described in the paper `Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift `__ . .. math:: y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta The mean and standard-deviation are calculated per-dimension over the mini-batches and :math:`\gamma` and :math:`\beta` are learnable parameter vectors of size `C` (where `C` is the input size). By default, the elements of :math:`\gamma` are set to 1 and the elements of :math:`\beta` are set to 0. At train time in the forward pass, the standard-deviation is calculated via the biased estimator, equivalent to ``torch.var(input, correction=0)``. However, the value stored in the moving average of the standard-deviation is calculated via the unbiased estimator, equivalent to ``torch.var(input, correction=1)``. Also by default, during training this layer keeps running estimates of its computed mean and variance, which are then used for normalization during evaluation. The running estimates are kept with a default :attr:`momentum` of 0.1. If :attr:`track_running_stats` is set to ``False``, this layer then does not keep running estimates, and batch statistics are instead used during evaluation time as well. .. note:: This :attr:`momentum` argument is different from one used in optimizer classes and the conventional notion of momentum. Mathematically, the update rule for running statistics here is :math:`\hat{x}_\text{new} = (1 - \text{momentum}) \times \hat{x} + \text{momentum} \times x_t`, where :math:`\hat{x}` is the estimated statistic and :math:`x_t` is the new observed value. Because the Batch Normalization is done over the `C` dimension, computing statistics on `(N, D, H, W)` slices, it's common terminology to call this Volumetric Batch Normalization or Spatio-temporal Batch Normalization. Args: num_features: :math:`C` from an expected input of size :math:`(N, C, D, H, W)` eps: a value added to the denominator for numerical stability. Default: 1e-5 momentum: the value used for the running_mean and running_var computation. Can be set to ``None`` for cumulative moving average (i.e. simple average). Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters. Default: ``True`` track_running_stats: a boolean value that when set to ``True``, this module tracks the running mean and variance, and when set to ``False``, this module does not track such statistics, and initializes statistics buffers :attr:`running_mean` and :attr:`running_var` as ``None``. When these buffers are ``None``, this module always uses batch statistics. in both training and eval modes. Default: ``True`` bias: If set to ``False``, the layer will not learn an additive bias (only relevant if :attr:`affine` is ``True``). Default: ``True`` Shape: - Input: :math:`(N, C, D, H, W)` - Output: :math:`(N, C, D, H, W)` (same shape as input) Examples:: >>> # With Learnable Parameters >>> m = nn.BatchNorm3d(100) >>> # Without Learnable Parameters >>> m = nn.BatchNorm3d(100, affine=False) >>> input = torch.randn(20, 100, 35, 45, 10) >>> output = m(input) r Nc|dkr%td|ddSNzexpected 5D input (got rrrOs r0rQzBatchNorm3d._check_input_dimarr2rerr+r2r0rrrr2rceZdZdZeZddZdS)raA :class:`torch.nn.BatchNorm3d` module with lazy initialization. Lazy initialization is done for the ``num_features`` argument of the :class:`BatchNorm3d` that is inferred from the ``input.size(1)``. The attributes that will be lazily initialized are `weight`, `bias`, `running_mean` and `running_var`. Check the :class:`torch.nn.modules.lazy.LazyModuleMixin` for further documentation on lazy modules and their limitations. Args: eps: a value added to the denominator for numerical stability. Default: 1e-5 momentum: the value used for the running_mean and running_var computation. Can be set to ``None`` for cumulative moving average (i.e. simple average). Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters. Default: ``True`` track_running_stats: a boolean value that when set to ``True``, this module tracks the running mean and variance, and when set to ``False``, this module does not track such statistics, and initializes statistics buffers :attr:`running_mean` and :attr:`running_var` as ``None``. When these buffers are ``None``, this module always uses batch statistics. in both training and eval modes. Default: ``True`` bias: If set to ``False``, the layer will not learn an additive bias (only relevant if :attr:`affine` is ``True``). Default: ``True`` r Nc|dkr%td|ddSrrrOs r0rQz LazyBatchNorm3d._check_input_dimrr2re)rfrgrhrirrrQr+r2r0rrfrr2rceZdZdZ ddddeded edzd ed ed edzd eddffdZddZ ddZ de de fdZ e ddZxZS)r aApplies Batch Normalization over a N-Dimensional input. The N-D input is a mini-batch of [N-2]D inputs with additional channel dimension as described in the paper `Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift `__ . .. math:: y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta The mean and standard-deviation are calculated per-dimension over all mini-batches of the same process groups. :math:`\gamma` and :math:`\beta` are learnable parameter vectors of size `C` (where `C` is the input size). By default, the elements of :math:`\gamma` are sampled from :math:`\mathcal{U}(0, 1)` and the elements of :math:`\beta` are set to 0. The standard-deviation is calculated via the biased estimator, equivalent to `torch.var(input, correction=0)`. Also by default, during training this layer keeps running estimates of its computed mean and variance, which are then used for normalization during evaluation. The running estimates are kept with a default :attr:`momentum` of 0.1. If :attr:`track_running_stats` is set to ``False``, this layer then does not keep running estimates, and batch statistics are instead used during evaluation time as well. .. note:: This :attr:`momentum` argument is different from one used in optimizer classes and the conventional notion of momentum. Mathematically, the update rule for running statistics here is :math:`\hat{x}_\text{new} = (1 - \text{momentum}) \times \hat{x} + \text{momentum} \times x_t`, where :math:`\hat{x}` is the estimated statistic and :math:`x_t` is the new observed value. Because the Batch Normalization is done for each channel in the ``C`` dimension, computing statistics on ``(N, +)`` slices, it's common terminology to call this Volumetric Batch Normalization or Spatio-temporal Batch Normalization. Currently :class:`SyncBatchNorm` only supports :class:`~torch.nn.DistributedDataParallel` (DDP) with single GPU per process. Use :meth:`torch.nn.SyncBatchNorm.convert_sync_batchnorm()` to convert :attr:`BatchNorm*D` layer to :class:`SyncBatchNorm` before wrapping Network with DDP. Args: num_features: :math:`C` from an expected input of size :math:`(N, C, +)` eps: a value added to the denominator for numerical stability. Default: ``1e-5`` momentum: the value used for the running_mean and running_var computation. Can be set to ``None`` for cumulative moving average (i.e. simple average). Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters. Default: ``True`` track_running_stats: a boolean value that when set to ``True``, this module tracks the running mean and variance, and when set to ``False``, this module does not track such statistics, and initializes statistics buffers :attr:`running_mean` and :attr:`running_var` as ``None``. When these buffers are ``None``, this module always uses batch statistics. in both training and eval modes. Default: ``True`` process_group: synchronization of stats happen within each process group individually. Default behavior is synchronization across the whole world bias: If set to ``False``, the layer will not learn an additive bias (only relevant if :attr:`affine` is ``True``). Default: ``True`` Shape: - Input: :math:`(N, C, +)` - Output: :math:`(N, C, +)` (same shape as input) .. note:: Synchronization of batchnorm statistics occurs only while training, i.e. synchronization is disabled when ``model.eval()`` is set or if ``self.training`` is otherwise ``False``. Examples:: >>> # xdoctest: +SKIP >>> # With Learnable Parameters >>> m = nn.SyncBatchNorm(100) >>> # creating process group (optional) >>> # ranks is a list of int identifying rank ids. >>> ranks = list(range(8)) >>> r1, r2 = ranks[:4], ranks[4:] >>> # Note: every rank calls into new_group for every >>> # process group created, even if that rank is not >>> # part of the group. >>> process_groups = [torch.distributed.new_group(pids) for pids in [r1, r2]] >>> process_group = process_groups[0 if dist.get_rank() <= 3 else 1] >>> # Without Learnable Parameters >>> m = nn.BatchNorm3d(100, affine=False, process_group=process_group) >>> input = torch.randn(20, 100, 35, 45, 10) >>> output = m(input) >>> # network is nn.BatchNorm layer >>> sync_bn_network = nn.SyncBatchNorm.convert_sync_batchnorm(network, process_group) >>> # only single gpu per process is currently supported >>> ddp_sync_bn_network = torch.nn.parallel.DistributedDataParallel( >>> sync_bn_network, >>> device_ids=[args.local_rank], >>> output_device=args.local_rank) rrTNrrrrrr process_grouprr c d||d} tj|||||fi| d| i||_dSru)r4r5r) rArrrrrrr#r$rrBrCs r0r5zSyncBatchNorm.__init__so%+U;;               +r2c|dkr%td|ddS)Nrz expected at least 2D input (got rrrOs r0rQzSyncBatchNorm._check_input_dim s< 99;;??U UUUVV V ?r2cV|ddkrtddS)Nr rz9SyncBatchNorm number of input channels should be non-zero)sizerrOs r0_check_non_zero_input_channelsz,SyncBatchNorm._check_non_zero_input_channelss4 ::a==A  K   r2rPc |||||jd}n|j}|jrb|jr[|jt d|jd|jd|jz }n|j} |jrd}n|j duo|j du} |jr|jr|j nd}|jr|jr|j nd}|oB|jo;tj otj }|r|jjddd tjfvr.t'd tjtj jj}|jr|j}tj |}|dk}|s*t1j||||j|j|||jS|st d t;j||j|j|||j||| S) z( Runs the forward pass. Nrwz$num_batches_tracked must not be Noner rxTcudahpuxpuz;SyncBatchNorm expected input tensor to be on GPU or XPU or zbn_training must be True)rQrrryrr(rrzitemr&r'r6 distributed is_availableis_initializedr#type_C_get_privateuse1_backend_namergroupWORLDrget_world_sizer{r|r%rrsync_batch_normapply) rArPr}r~r&r' need_syncr world_sizes r0rzSyncBatchNorm.forwards e$$$ ++E222 = ), & &)- & = ;T5 ;'/$%KLLL  $ ) )! , , ,}$-043K3P3P3R3R-R**-1]*  = UKK,4T4;Kt;SK &*] Xd6N XD  TX %)M WT5M WD  SW   3  3!..00 3!0022   '| 6688 ) !Bx==??BB "-39M! 3 $ 2 *99-HHJ"QI <  *    A$%?@@@"(  *   r2c |}t|tjjjjrtj|j|j|j |j |j ||j du}|j rCtj 5|j|_|j |_ dddn #1swxYwY|j|_|j|_|j|_|j|_t'|dr |j|_|D]/\}}|||||0~|S)aaConverts all :attr:`BatchNorm*D` layers in the model to :class:`torch.nn.SyncBatchNorm` layers. Args: module (nn.Module): module containing one or more :attr:`BatchNorm*D` layers process_group (optional): process group to scope synchronization, default is the whole world Returns: The original :attr:`module` with the converted :class:`torch.nn.SyncBatchNorm` layers. If the original :attr:`module` is a :attr:`BatchNorm*D` layer, a new :class:`torch.nn.SyncBatchNorm` layer object will be returned instead. Example:: >>> # Network with nn.BatchNorm layer >>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_CUDA) >>> module = torch.nn.Sequential( >>> torch.nn.Linear(20, 100), >>> torch.nn.BatchNorm1d(100), >>> ).cuda() >>> # creating process group (optional) >>> # ranks is a list of int identifying rank ids. >>> ranks = list(range(8)) >>> r1, r2 = ranks[:4], ranks[4:] >>> # Note: every rank calls into new_group for every >>> # process group created, even if that rank is not >>> # part of the group. >>> # xdoctest: +SKIP("distributed") >>> process_groups = [torch.distributed.new_group(pids) for pids in [r1, r2]] >>> process_group = process_groups[0 if dist.get_rank() <= 3 else 1] >>> sync_bn_module = torch.nn.SyncBatchNorm.convert_sync_batchnorm(module, process_group) Nrqconfig)rr6nnmodules batchnormrsr rrrrrrno_gradr%r&r'r(ryhasattrrnamed_children add_moduleconvert_sync_batchnorm)clsmoduler module_outputnamechilds r0rz$SyncBatchNorm.convert_sync_batchnormysH feh.8C D D 7!H22#  *[,3M} 5]__55+1=M()/M&555555555555555*0)rs ********UUUUUUUUUU888888!!!!!!   | | | | | | | | ~H H H H H H H H VL$L$L$L$L$OYL$L$L$^KTKTKTKTKT*KTKTKT\!T!T!T!T!TmZ!T!T!THLNLNLNLNLN*LNLNLN^!N!N!N!N!NmZ!N!N!NHLNLNLNLNLN*LNLNLN^!N!N!N!N!NmZ!N!N!NHmmmmmJmmmmmr2