3jSSKJr SSKrSSKJr SSKJr SS Sjjr"SS\R5rg) ) annotationsN)nn)metricsc [R"XX#XF5n[R"SU- S- SSS9nUS:Xa[R"U5nU$US:Xa[R "U5nU$US:XaU$[ S 5e) aCompute a loss based on the SSIM measurement. The loss, or the Structural dissimilarity (DSSIM) is described as: .. math:: \text{loss}(x, y) = \frac{1 - \text{SSIM}(x, y)}{2} See :meth:`~kornia.losses.ssim` for details about SSIM. Args: img1: the first input image with shape :math:`(B, C, H, W)`. img2: the second input image with shape :math:`(B, C, H, W)`. window_size: the size of the gaussian kernel to smooth the images. max_val: the dynamic range of the images. eps: Small value for numerically stability when dividing. reduction : Specifies the reduction to apply to the output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction will be applied, ``'mean'``: the sum of the output will be divided by the number of elements in the output, ``'sum'``: the output will be summed. padding: ``'same'`` | ``'valid'``. Whether to only use the "valid" convolution area to compute SSIM to match the MATLAB implementation of original SSIM paper. Returns: The loss based on the ssim index. Examples: >>> input1 = torch.rand(1, 4, 5, 5) >>> input2 = torch.rand(1, 4, 5, 5) >>> loss = ssim_loss(input1, input2, 5) ?r)minmaxmeansumnonezInvalid reduction option.)rssimtorchclampr r NotImplementedError) img1img2 window_sizemax_valeps reductionpaddingssim_maplosss L/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/losses/ssim.py ssim_lossrsT%\\$kCYH ;;h!+ :DFzz$ K e yy K f   K""=>>cV^\rSrSrSrSSU4SjjjrSSjrSrU=r$) SSIMLossUaaCreate a criterion that computes a loss based on the SSIM measurement. The loss, or the Structural dissimilarity (DSSIM) is described as: .. math:: \text{loss}(x, y) = \frac{1 - \text{SSIM}(x, y)}{2} See :meth:`~kornia.losses.ssim_loss` for details about SSIM. Args: window_size: the size of the gaussian kernel to smooth the images. max_val: the dynamic range of the images. eps: Small value for numerically stability when dividing. reduction : Specifies the reduction to apply to the output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction will be applied, ``'mean'``: the sum of the output will be divided by the number of elements in the output, ``'sum'``: the output will be summed. padding: ``'same'`` | ``'valid'``. Whether to only use the "valid" convolution area to compute SSIM to match the MATLAB implementation of original SSIM paper. Returns: The loss based on the ssim index. Examples: >>> input1 = torch.rand(1, 4, 5, 5) >>> input2 = torch.rand(1, 4, 5, 5) >>> criterion = SSIMLoss(5) >>> loss = criterion(input1, input2) c^>[TU]5 XlX lX0lX@lXPlgN)super__init__rrrrr)selfrrrrr __class__s rr%SSIMLoss.__init__vs*  +% '# rc [XURURURURUR 5$r#)rrrrrr)r&rrs rforwardSSIMLoss.forwards0T%5%5t||TXXt~~_c_k_kllr)rrrrrrg-q=r same) rintrfloatrr/rstrrr0returnNone)r torch.Tensorrr3r1r3) __name__ __module__ __qualname____firstlineno____doc__r%r*__static_attributes__ __classcell__)r's@rr r UsVBsy$$).$;@$UX$lo$ $$mmrr r,)rr3rr3rr.rr/rr/rr0rr0r1r3) __future__rrrkorniarrModuler rrr?s$# 8 8 88 8  8  888v,mryy,mr