o ϯi-@sddlmZmZddlZddlmZddlmZGdddejZ dddd d d d d d d ej d d ej f de de dede de de de de de dede de dedejfddZdS))CallableOptionalN)ResBlockcsReZdZdZ  d dedeedeeffdd Zd ejd ejfd d Z Z S)SqueezeAndExcitationLayer2DzL2D Squeeze and excitation layer, as per https://arxiv.org/pdf/1709.01507.pdfN in_planesreduction_ratioreduced_planesc sttd|_t|t|ksJd|dur||n|}ttj||ddddttj||ddddt |_ dS)a^ Args: in_planes (int): input channel dimension. reduction_ratio (int): factor by which in_planes should be reduced to get the output channel dimension. reduced_planes (int): Output channel dimension. Only one of reduction_ratio or reduced_planes should be defined. )r zHOnly of reduction_ratio or reduced_planes should be defined for SE layerNr T kernel_sizestridebias) super__init__nnAdaptiveAvgPool2davgpoolbool SequentialConv2dReLUSigmoid excitation)selfrrr  __class__Z/home/ubuntu/.local/lib/python3.10/site-packages/pytorchvideo/layers/squeeze_excitation.pyr s   z$SqueezeAndExcitationLayer2D.__init__xreturncCs ||}||}||}|S)zL Args: x (tensor): 2D image of format C * H * W )rr)rr x_squeezed x_excitedx_scaledrrrforward.s  z#SqueezeAndExcitationLayer2D.forward)rN) __name__ __module__ __qualname____doc__intrrtorchTensorr$ __classcell__rrrrr s!rFrcCs||S)Nr)ryrrr>sr.r gh㈵>g?dim_indim_out branch_fusionconv_a_kernel_size conv_a_strideconv_a_paddingconv_b_kernel_size conv_b_strideconv_b_paddingnormnorm_eps norm_momentum activationr c Cstj|||||dd| || | |r|nttj|||| | dd| || | g}|r3|t||dtj|}d\}}|| dksF||krXtj||d|| dd}| || | }t||||rd||dSd|dS) ur 2-D Residual block with squeeze excitation (SE2D) for 2d. Performs a summation between an identity shortcut in branch1 and a main block in branch2. When the input and output dimensions are different, a convolution followed by a normalization will be performed. :: Input |-------+ ↓ | conv2d | ↓ | Norm | ↓ | activation | ↓ | conv2d | ↓ | Norm | ↓ | SE2D | ↓ } Summation ←-+ ↓ Activation Normalization examples include: BatchNorm3d and None (no normalization). Activation examples include: ReLU, Softmax, Sigmoid, and None (no activation). Transform examples include: BottleneckBlock. Args: dim_in (int): input channel size to the bottleneck block. dim_out (int): output channel size of the bottleneck. use_se (bool): if true, use squeeze excitation layer in the bottleneck. se_reduction_ratio (int): factor by which input channels should be reduced to get the output channel dimension in SE layer. branch_fusion (callable): a callable that constructs summation layer. Examples include: lambda x, y: x + y, OctaveSum. conv_a_kernel_size (tuple): convolutional kernel size(s) for conv_a. conv_a_stride (tuple): convolutional stride size(s) for conv_a. conv_a_padding (tuple): convolutional padding(s) for conv_a. conv_b_kernel_size (tuple): convolutional kernel size(s) for conv_b. conv_b_stride (tuple): convolutional stride size(s) for conv_b. conv_b_padding (tuple): convolutional padding(s) for conv_b. norm (callable): a callable that constructs normalization layer. Examples include nn.BatchNorm3d, None (not performing normalization). norm_eps (float): normalization epsilon. norm_momentum (float): normalization momentum. activation (callable): a callable that constructs activation layer in bottleneck and block. Examples include: nn.ReLU, nn.Softmax, nn.Sigmoid, and None (not performing activation). Returns: (nn.Module): resnet basic block layer. F)r r paddingr)r)NNr r N) branch1_conv branch1_normbranch2r<r2)rrIdentityappendrrr)r0r1use_sese_reduction_ratior2r3r4r5r6r7r8r9r:r;r<r@r>r?rrr(create_audio_2d_squeeze_excitation_block9sZR     rE)typingrrr*torch.nnrpytorchvideo.models.resnetrModuler BatchNorm2drr)floatrErrrrs^  2