o ki@sddlmZddlZddlmZddlmZmZddlm Z m Z m Z ddl m ZddlmZdd lmZgd ZGd d d eZGd ddeZGdddeeZGdddeZGdddeeZGdddeZGdddeeZGdddeZGdddeeZGdddeZ dS))AnyN)Tensor) functionalinit) ParameterUninitializedBufferUninitializedParameter) SyncBatchNorm)LazyModuleMixin)Module) BatchNorm1dLazyBatchNorm1d BatchNorm2dLazyBatchNorm2d BatchNorm3dLazyBatchNorm3dr c seZdZUdZdZgdZeed<eed<edBed<e ed<e ed <  ddedededBde d e d df fdd Z dddZ dddZ ddZ ddZ dfdd ZZS) _NormBasez,Common base of _InstanceNorm and _BatchNorm.)track_running_statsmomentumeps num_featuresaffinerrNrrrh㈵>皙?Treturnc s(||d}t||_||_||_||_||_|jr5ttj |fi||_ ttj |fi||_ n | dd| dd|jr|| dtj|fi|| dtj|fi|||| dtj d dtjid d |D|n| dd| dd| dd|dS) Ndevicedtypeweightbias running_mean running_varnum_batches_trackedrrcSi|] \}}|dkr||qSr.0kvr'r'd/var/www/addictedbytheproject.nl/epg/venv/lib/python3.10/site-packages/torch/nn/modules/batchnorm.py Lz&_NormBase.__init__..r)super__init__rrrrrrtorchemptyr r!register_parameterregister_bufferzerosonestensorlongitemsreset_parameters selfrrrrrrrfactory_kwargs __class__r'r,r1&sH       z_NormBase.__init__cCs.|jr|j|jd|jdSdS)Nr )rr"zero_r#fill_r$r=r'r'r,reset_running_statsVs   z_NormBase.reset_running_statscCs.||jrt|jt|jdSdSN)rDrrones_r zeros_r!rCr'r'r,r;^s  z_NormBase.reset_parameterscCstrE)NotImplementedErrorr=inputr'r'r,_check_input_dimdsz_NormBase._check_input_dimcCsdjdi|jS)Nzj{num_features}, eps={eps}, momentum={momentum}, affine={affine}, track_running_stats={track_running_stats}r')format__dict__rCr'r'r, extra_reprgs z_NormBase.extra_reprc s|dd}|dus|dkr4|jr4|d} | |vr4|jdur*|jjtdkr*|jntjdtjd|| <t|||||||dS)Nversionrr$metarr&) getrr$rr2r8r9r0_load_from_state_dict) r= state_dictprefixlocal_metadatastrict missing_keysunexpected_keys error_msgsrOnum_batches_tracked_keyr?r'r,rRms$  z_NormBase._load_from_state_dictrrTTNNrN)__name__ __module__ __qualname____doc___version __constants__int__annotations__floatboolr1rDr;rKrNrR __classcell__r'r'r?r,rsF    0  rc sZeZdZ      ddedededBded ed df fd d Zd ed efddZZ S) _BatchNormrrTNrrrrrrc s*||d}tj|||||fi|dSNr)r0r1r<r?r'r,r1s   z_BatchNorm.__init__rJc Cs|||jdur d}n|j}|jr1|jr1|jdur1|jd|jdur.dt|j}n|j} |jr8d}n |jduoA|jdu} t ||jrL|jrO|jnd|jrV|jrY|jnd|j |j |||j S)Nr ?T)rKrtrainingrr$add_rer"r#F batch_normr r!r)r=rJexponential_average_factor bn_trainingr'r'r,forwards:      z_BatchNorm.forwardr[) r]r^r_rcrerfr1rrrrgr'r'r?r,rhs* rhcsXeZdZUeed<eed<      d dfdd Zdfd d Zdd d ZZS) _LazyNormBaser r!rrTNrcs||d}tjd||ddfi|||_||_|jr,tdi||_tdi||_|jrUtdi||_tdi||_ t j ddt j idd| D|_dSdS) NrrFrcSr%r&r'r(r'r'r,r-r.z*_LazyNormBase.__init__..r'r/)r0r1rrrr r!rr"r#r2r8r9r:r$)r=rrrrrrr>r?r'r,r1s4   z_LazyNormBase.__init__cs(|s|jdkrtdSdSdS)Nr)has_uninitialized_paramsrr0r;rCr?r'r,r;sz_LazyNormBase.reset_parameterscCs|rJ|jd|_|jr1t|jtstdt|jts!td|j |jf|j |jf|j rD|j |jf|j |jf| dSdS)Nr z-self.weight must be an UninitializedParameterz+self.bias must be an UninitializedParameter)rtshaperr isinstancer rAssertionErrorr! materializerr"r#r;rIr'r'r,initialize_parameterss(    z#_LazyNormBase.initialize_parametersr[r\) r]r^r_rrdr1r;ryrgr'r'r?r,rss (rsc@eZdZdZdddZdS)r a Applies 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`` 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) rNcC4|dkr|dkrtd|ddSdSNrzexpected 2D or 3D input (got D input)dim ValueErrorrIr'r'r,rKbzBatchNorm1d._check_input_dimr\r]r^r_r`rKr'r'r'r,r sFr c@eZdZdZeZdddZdS)raRA :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`` rNcCr{r|rrIr'r'r,rKrz LazyBatchNorm1d._check_input_dimr\)r]r^r_r`r cls_to_becomerKr'r'r'r,rgrc@rz)ra Applies 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`` 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) rNcC$|dkrtd|ddSNzexpected 4D input (got r~rrIr'r'r,rK zBatchNorm2d._check_input_dimr\rr'r'r'r,rGrc@r)raUA :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`` rNcCrrrrIr'r'r,rKrz LazyBatchNorm2d._check_input_dimr\)r]r^r_r`rrrKr'r'r'r,rrrc@rz)ra Applies 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`` 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) rNcCrNzexpected 5D input (got r~rrIr'r'r,rK@rzBatchNorm3d._check_input_dimr\rr'r'r'r,rrrc@r)raUA :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`` rNcCrrrrIr'r'r,rKbrz LazyBatchNorm3d._check_input_dimr\)r]r^r_r`rrrKr'r'r'r,rErrcseZdZdZ       ddedededBd ed ed edBd dffd d ZdddZ dddZ de d e fddZ e dddZZS)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 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) rrTNrrrrr process_grouprc s0||d} tj|||||fi| ||_dSri)r0r1r) r=rrrrrrrrr>r?r'r,r1s   zSyncBatchNorm.__init__cCs$|dkrtd|ddS)Nrz expected at least 2D input (got r~rrIr'r'r,rKrzSyncBatchNorm._check_input_dimcCs|ddkr tddS)Nr rz9SyncBatchNorm number of input channels should be non-zero)sizerrIr'r'r,_check_non_zero_input_channelss z,SyncBatchNorm._check_non_zero_input_channelsrJc Cs|||||jdurd}n|j}|jr:|jr:|jdur$td|jd|jdur7d|j}n|j} |jrAd}n |j duoJ|j du} |jrR|jrU|j nd}|jr]|jr`|j nd}|op|jopt j opt j }|r|jjddd t jfvrtd t jt j jj}|jr|j}t j |}|dk}|st||||j|j|||jS|std t||j|j|||j||| S) z( Runs the forward pass. Nrjz$num_batches_tracked must not be Noner rkTcudahpuxpuz;SyncBatchNorm expected input tensor to be on GPU or XPU or zbn_training must be True)rKrrrlrr$rwrmitemr"r#r2 distributed is_availableis_initializedrtype_C_get_privateuse1_backend_namergroupWORLDrget_world_sizernror r!rsync_batch_normapply) r=rJrprqr"r# need_syncr world_sizer'r'r,rrs          zSyncBatchNorm.forwardcCs|}t|tjjjjrStj|j|j|j |j |j |}|j r:t |j |_ |j|_Wdn1s5wY|j|_|j|_|j|_|j|_t|drS|j|_|D]\}}|||||qW~|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) Nqconfig)rvr2nnmodules batchnormrhr rrrrrno_gradr r!r"r#r$rlhasattrrnamed_children add_moduleconvert_sync_batchnorm)clsmoduler module_outputnamechildr'r'r,rLs6$    z$SyncBatchNorm.convert_sync_batchnorm)rrTTNNNr\rE)r]r^r_r`rcrerfrr1rKrrrr classmethodrrgr'r'r?r,r gs:i   cr )typingrr2rtorch.nnrrnrtorch.nn.parameterrrr _functionsr rlazyr rr __all__rrhrsr rrrrrr'r'r'r,s&      wCHL"M"M"