Sj TddlZddlmcmZddlmZddlmZm Z gdZ Gdde Z Gdd e Z Gd d ee Z Gd d e ZGddee ZGdde ZGddee Zy)N)Tensor) _LazyNormBase _NormBase)InstanceNorm1dInstanceNorm2dInstanceNorm3dLazyInstanceNorm1dLazyInstanceNorm2dLazyInstanceNorm3dceZdZ ddddededededed ed dffd Zd Zd ZdZ dZ dfd Z de d e fdZ xZS) _InstanceNormNT)bias num_featuresepsmomentumaffinetrack_running_statsrreturnc@||d} t ||||||fi| d|iy)N)devicedtyper)super__init__) selfrrrrrrrrfactory_kwargs __class__s \/var/www/html/ia-enhanced/venv/lib/python3.12/site-packages/torch/nn/modules/instancenorm.pyrz_InstanceNorm.__init__s?%+U;           ctNNotImplementedErrorrinputs r_check_input_dimz_InstanceNorm._check_input_dim-!!rctr!r"rs r_get_no_batch_dimz_InstanceNorm._get_no_batch_dim0r'rc`|j|jdjdS)Nr)_apply_instance_norm unsqueezesqueezer$s r_handle_no_batch_inputz$_InstanceNorm._handle_no_batch_input3s'(();<DDQGGrc tj||j|j|j|j |j xs |j |j|j|jSd|jS)Ng) F instance_norm running_mean running_varweightrtrainingrrrr$s rr,z"_InstanceNorm._apply_instance_norm6sy        KK II MM 9!9!99!]]6DMM HH  =@ HH  rc |jdd}||jsg} dD]} || z} | |vs | j| t| dkDrd|jdj dj d| D|j j| D]} |j| t |)|||||||y)Nversion)r3r4raUnexpected running stats buffer(s) {names} for {klass} with track_running_stats=False. If state_dict is a checkpoint saved before 0.4.0, this may be expected because {klass} does not track running stats by default since 0.4.0. Please remove these keys from state_dict. If the running stats are actually needed, instead set track_running_stats=True in {klass} to enable them. See the documentation of {klass} for details.z and c3(K|] }d|d yw)"N).0ks r z6_InstanceNorm._load_from_state_dict.._s*PQqc8*Ps)namesklass) getrappendlenformatjoinr__name__popr_load_from_state_dict) r state_dictprefixlocal_metadatastrict missing_keysunexpected_keys error_msgsr8running_stats_keysnamekeyrs rrHz#_InstanceNorm._load_from_state_dictBs!$$Y5 ?4#;#;!# 7 3tm*$&--c2 3%&*!!@AG%ll*P=O*PP"nn55AGA .(CNN3'( %        rr%c |j||j|jz }|j||jk7rU|j r.t d|d|jd|j|dtjd|dd|j|jk(r|j|S|j|S) Nzexpected input's size at dim=z to match num_features (z ), but got: .zinput's size at dim=z does not match num_features. You can silence this warning by not passing in num_features, which is not used because affine=False) stacklevel) r&dimr*sizerr ValueErrorwarningswarnr/r,)rr% feature_dims rforwardz_InstanceNorm.forwardps e$iikD$:$:$<< ::k "d&7&7 7{{ 3K=A**+< ;8O7PPQS  *;-8== !  99;$002 2..u5 5((//r)gh㈵>g?FFNNrN)rF __module__ __qualname__intfloatboolrr&r*r/r,rHrr] __classcell__)rs@rrrs$)        "     .""H  ,  , \0V00rrc$eZdZdZdefdZddZy)ra Applies Instance Normalization. This operation applies Instance Normalization over a 2D (unbatched) or 3D (batched) input as described in the paper `Instance Normalization: The Missing Ingredient for Fast Stylization `__. .. math:: y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta The mean and standard-deviation are calculated per-dimension separately for each object in a mini-batch. :math:`\gamma` and :math:`\beta` are learnable parameter vectors of size `C` (where `C` is the number of features or channels of the input) if :attr:`affine` is ``True``. The variance is calculated via the biased estimator, equivalent to `torch.var(input, correction=0)`. By default, this layer uses instance statistics computed from input data in both training and evaluation modes. If :attr:`track_running_stats` is set to ``True``, 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. .. 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. .. note:: :class:`InstanceNorm1d` and :class:`LayerNorm` are very similar, but have some subtle differences. :class:`InstanceNorm1d` is applied on each channel of channeled data like multidimensional time series, but :class:`LayerNorm` is usually applied on entire sample and often in NLP tasks. Additionally, :class:`LayerNorm` applies elementwise affine transform, while :class:`InstanceNorm1d` usually don't apply affine transform. 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. Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters, initialized the same way as done for batch normalization. Default: ``False`` 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 always uses batch statistics in both training and eval modes. Default: ``False`` 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, L)` or :math:`(C, L)` - Output: :math:`(N, C, L)` or :math:`(C, L)` (same shape as input) Examples:: >>> # Without Learnable Parameters >>> m = nn.InstanceNorm1d(100) >>> # With Learnable Parameters >>> m = nn.InstanceNorm1d(100, affine=True) >>> input = torch.randn(20, 100, 40) >>> output = m(input) rcyNrUr;r)s rr*z InstanceNorm1d._get_no_batch_dimrNcb|jdvrtd|jdyN)rUzexpected 2D or 3D input (got D input)rWrYr$s rr&zInstanceNorm1d._check_input_dim0 99;f $`__. .. math:: y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta The mean and standard-deviation are calculated per-dimension separately for each object in a mini-batch. :math:`\gamma` and :math:`\beta` are learnable parameter vectors of size `C` (where `C` is the input size) if :attr:`affine` is ``True``. The standard-deviation is calculated via the biased estimator, equivalent to `torch.var(input, correction=0)`. By default, this layer uses instance statistics computed from input data in both training and evaluation modes. If :attr:`track_running_stats` is set to ``True``, 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. .. 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. .. note:: :class:`InstanceNorm2d` and :class:`LayerNorm` are very similar, but have some subtle differences. :class:`InstanceNorm2d` is applied on each channel of channeled data like RGB images, but :class:`LayerNorm` is usually applied on entire sample and often in NLP tasks. Additionally, :class:`LayerNorm` applies elementwise affine transform, while :class:`InstanceNorm2d` usually don't apply affine transform. Args: num_features: :math:`C` from an expected input of size :math:`(N, C, H, W)` or :math:`(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. Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters, initialized the same way as done for batch normalization. Default: ``False`` 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 always uses batch statistics in both training and eval modes. Default: ``False`` 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)` or :math:`(C, H, W)` - Output: :math:`(N, C, H, W)` or :math:`(C, H, W)` (same shape as input) Examples:: >>> # Without Learnable Parameters >>> m = nn.InstanceNorm2d(100) >>> # With Learnable Parameters >>> m = nn.InstanceNorm2d(100, affine=True) >>> input = torch.randn(20, 100, 35, 45) >>> output = m(input) rcyNrkr;r)s rr*z InstanceNorm2d._get_no_batch_dimGrhrNcb|jdvrtd|jdyN)rkzexpected 3D or 4D input (got rlrmr$s rr&zInstanceNorm2d._check_input_dimJrnrr^ror;rrrrsFP3Trrc(eZdZdZeZdefdZddZy)r a3A :class:`torch.nn.InstanceNorm2d` module with lazy initialization of the ``num_features`` argument. The ``num_features`` argument of the :class:`InstanceNorm2d` 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: num_features: :math:`C` from an expected input of size :math:`(N, C, H, W)` or :math:`(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. Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters, initialized the same way as done for batch normalization. Default: ``False`` 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 always uses batch statistics in both training and eval modes. Default: ``False`` 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)` or :math:`(C, H, W)` - Output: :math:`(N, C, H, W)` or :math:`(C, H, W)` (same shape as input) rcyrwr;r)s rr*z$LazyInstanceNorm2d._get_no_batch_dimorhrNcb|jdvrtd|jdyryrmr$s rr&z#LazyInstanceNorm2d._check_input_dimrrnrr^) rFr_r`rprrtrar*r&r;rrr r O :#M3Trr c$eZdZdZdefdZddZy)r a Applies Instance Normalization. This operation applies Instance Normalization over a 5D input (a mini-batch of 3D inputs with additional channel dimension) as described in the paper `Instance Normalization: The Missing Ingredient for Fast Stylization `__. .. math:: y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta The mean and standard-deviation are calculated per-dimension separately for each object in a mini-batch. :math:`\gamma` and :math:`\beta` are learnable parameter vectors of size C (where C is the input size) if :attr:`affine` is ``True``. The standard-deviation is calculated via the biased estimator, equivalent to `torch.var(input, correction=0)`. By default, this layer uses instance statistics computed from input data in both training and evaluation modes. If :attr:`track_running_stats` is set to ``True``, 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. .. 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. .. note:: :class:`InstanceNorm3d` and :class:`LayerNorm` are very similar, but have some subtle differences. :class:`InstanceNorm3d` is applied on each channel of channeled data like 3D models with RGB color, but :class:`LayerNorm` is usually applied on entire sample and often in NLP tasks. Additionally, :class:`LayerNorm` applies elementwise affine transform, while :class:`InstanceNorm3d` usually don't apply affine transform. Args: num_features: :math:`C` from an expected input of size :math:`(N, C, D, H, W)` or :math:`(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. Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters, initialized the same way as done for batch normalization. Default: ``False`` 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 always uses batch statistics in both training and eval modes. Default: ``False`` 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)` or :math:`(C, D, H, W)` - Output: :math:`(N, C, D, H, W)` or :math:`(C, D, H, W)` (same shape as input) Examples:: >>> # Without Learnable Parameters >>> m = nn.InstanceNorm3d(100) >>> # With Learnable Parameters >>> m = nn.InstanceNorm3d(100, affine=True) >>> input = torch.randn(20, 100, 35, 45, 10) >>> output = m(input) rcyNrzr;r)s rr*z InstanceNorm3d._get_no_batch_dimrhrNcb|jdvrtd|jdyN)rzzexpected 4D or 5D input (got rlrmr$s rr&zInstanceNorm3d._check_input_dimrnrr^ror;rrr r wsEN3Trr c(eZdZdZeZdefdZddZy)r aEA :class:`torch.nn.InstanceNorm3d` module with lazy initialization of the ``num_features`` argument. The ``num_features`` argument of the :class:`InstanceNorm3d` 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: num_features: :math:`C` from an expected input of size :math:`(N, C, D, H, W)` or :math:`(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. Default: 0.1 affine: a boolean value that when set to ``True``, this module has learnable affine parameters, initialized the same way as done for batch normalization. Default: ``False`` 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 always uses batch statistics in both training and eval modes. Default: ``False`` 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)` or :math:`(C, D, H, W)` - Output: :math:`(N, C, D, H, W)` or :math:`(C, D, H, W)` (same shape as input) rcyrr;r)s rr*z$LazyInstanceNorm3d._get_no_batch_dimrhrNcb|jdvrtd|jdyrrmr$s rr&z#LazyInstanceNorm3d._check_input_dimrnrr^) rFr_r`rpr rtrar*r&r;rrr r r~rr )rZtorch.nn.functionalnn functionalr1torchr batchnormrr__all__rrr rr r r r;rrrs/ p0Ip0fLT]LT^$T $TNNT]NTb%T %TPMT]MT`%T %Tr