a þdˆã@sPddlmZmZmZddlZddlmZddlmmZ Gdd„dej ƒZ dS)é)ÚListÚOptionalÚTupleNc s´eZdZdZgd¢dddddfeeeeeefeeedd œ‡fd d „ Zde ee e j e e j e jd œd d„Zde ee e j e e j e jd œdd„Ze je je jdœdd„Z‡ZS)Ú MS_SSIMLossa€Creates a criterion that computes MSSIM + L1 loss. According to [1], we compute the MS_SSIM + L1 loss as follows: .. math:: \text{loss}(x, y) = \alpha \cdot \mathcal{L_{MSSIM}}(x,y)+(1 - \alpha) \cdot G_\alpha \cdot \mathcal{L_1}(x,y) Where: - :math:`\alpha` is the weight parameter. - :math:`x` and :math:`y` are the reconstructed and true reference images. - :math:`\mathcal{L_{MSSIM}}` is the MS-SSIM loss. - :math:`G_\alpha` is the sigma values for computing multi-scale SSIM. - :math:`\mathcal{L_1}` is the L1 loss. Reference: [1]: https://research.nvidia.com/sites/default/files/pubs/2017-03_Loss-Functions-for/NN_ImgProc.pdf#page11 Args: sigmas: gaussian sigma values. data_range: the range of the images. K: k values. alpha : specifies the alpha value compensation: specifies the scaling coefficient. reduction : Specifies the reduction to apply to the output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction will be applied, ``'mean'``: the sum of the output will be divided by the number of elements in the output, ``'sum'``: the output will be summed. Returns: The computed loss. Shape: - Input1: :math:`(N, C, H, W)`. - Input2: :math:`(N, C, H, W)`. - Output: :math:`(N, H, W)` or scalar if reduction is set to ``'mean'`` or ``'sum'``. Examples: >>> input1 = torch.rand(1, 3, 5, 5) >>> input2 = torch.rand(1, 3, 5, 5) >>> criterion = kornia.losses.MS_SSIMLoss() >>> loss = criterion(input1, input2) )gà?çð?g@g@g @r)g{®Gáz„?g¸…ëQ¸ž?gš™™™™™™?gi@ÚmeanN)ÚsigmasÚ data_rangeÚKÚalphaÚ compensationÚ reductionÚreturnc s tƒ ¡||_|d|d|_|d|d|_td|dƒ|_||_||_||_ td|ddƒ}t   dt |ƒd||f¡}t |ƒD]€\} } | || ¡|d| dddd…dd…f<| || ¡|d| dddd…dd…f<| || ¡|d| dddd…dd…f<qŽ| d|¡dS)NréééÿÿÿÿééÚ_g_masks)ÚsuperÚ__init__ÚDRÚC1ÚC2ÚintÚpadr r r ÚtorchÚzerosÚlenÚ enumerateÚ_fspecial_gauss_2dZregister_buffer) Úselfrr r r r r Z filter_sizeÚg_masksÚidxÚsigma©Ú __class__©ú^/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/kornia/losses/ms_ssim.pyr7s ((*zMS_SSIMLoss.__init__)Úsizer$ÚdeviceÚdtypercCsNtj|||d}||d8}t |d d|d¡}|| ¡}| d¡S)z½Create 1-D gauss kernel. Args: size: the size of gauss kernel. sigma: sigma of normal distribution. Returns: 1D kernel (size). )r*r+rr)rZarangeÚexpÚsumZreshape)r!r)r$r*r+ZcoordsÚgr'r'r(Ú_fspecial_gauss_1dUs   zMS_SSIMLoss._fspecial_gauss_1dcCs| ||||¡}t ||¡S)zÄCreate 2-D gauss kernel. Args: size: the size of gauss kernel. sigma: sigma of normal distribution. Returns: 2D kernel (size x size). )r/rÚouter)r!r)r$r*r+Z gaussian_vecr'r'r(r gs zMS_SSIMLoss._fspecial_gauss_2d)Úimg1Úimg2rcCspt|tjƒstdt|ƒ›ƒ‚t|tjƒss