a d&@sTddlZddlmZddlmZddlmZeeedddZGdddejZdS) N)Tensor)KORNIA_CHECK_SHAPEpredtargetreturncCst|gdt|gd|jddks:td|j|jdd|jddksntd|jd|j|j|jkstd |jd|j||jd d }||jd d }|j\}}d |d }d ||}|jddd\}} tj||jdd dd|d } || | d f} | ||} | j ddd} | | d} | d |  d}d | |}|dkr|dddf|ddd f|dddf<| | d }|S)aCriterion that computes a surrogate binary intersection-over-union (IoU) loss. According to [2], we compute the IoU as follows: .. math:: \text{IoU}(x, class) = \frac{|X \cap Y|}{|X \cup Y|} [1] approximates this fomular with a surrogate, which is fully differentable. Where: - :math:`X` expects to be the scores of each class. - :math:`Y` expects to be the binary tensor with the class labels. the loss, is finally computed as: .. math:: \text{loss}(x, class) = 1 - \text{IoU}(x, class) Reference: [1] http://proceedings.mlr.press/v37/yub15.pdf [2] https://arxiv.org/pdf/1705.08790.pdf . note:: This loss function only supports binary labels. For multi-class labels please use the Lovasz-Softmax loss. Args: pred: logits tensor with shape :math:`(N, 1, H, W)`. labels: labels tensor with shape :math:`(N, H, W)` with binary values. Return: a scalar with the computed loss. Example: >>> N = 1 # num_classes >>> pred = torch.randn(1, N, 3, 5, requires_grad=True) >>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(N) >>> output = lovasz_hinge_loss(pred, target) >>> output.backward() )B1HW)rr r z,Invalid pred shape, we expect Bx1xHxW. Got: Nz.pred and target shapes must be the same. Got: z and z1pred and target must be in the same device. Got: rg@g?T)ZdimZ descending)device)Zkeepdim.)rshape ValueErrorrZreshapesorttorchZarangerepeatviewsumZcumsumZrelumean)rrZ pred_flattenZtarget_flattenrNZsignserrorsZ errors_sortedZ permutationZ batch_indexZ target_sortedZtarget_sorted_sum intersectionunionZgradientZlossrc/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/kornia/losses/lovasz_hinge.pylovasz_hinge_loss s2+    &   ,rcs8eZdZdZddfdd ZeeedddZZS) LovaszHingeLossa"Criterion that computes a surrogate binary intersection-over-union (IoU) loss. According to [2], we compute the IoU as follows: .. math:: \text{IoU}(x, class) = \frac{|X \cap Y|}{|X \cup Y|} [1] approximates this fomular with a surrogate, which is fully differentable. Where: - :math:`X` expects to be the scores of each class. - :math:`Y` expects to be the binary tensor with the class labels. the loss, is finally computed as: .. math:: \text{loss}(x, class) = 1 - \text{IoU}(x, class) Reference: [1] http://proceedings.mlr.press/v37/yub15.pdf [2] https://arxiv.org/pdf/1705.08790.pdf . note:: This loss function only supports binary labels. For multi-class labels please use the Lovasz-Softmax loss. Args: pred: logits tensor with shape :math:`(N, 1, H, W)`. labels: labels tensor with shape :math:`(N, H, W)` with binary values. Return: a scalar with the computed loss. Example: >>> N = 1 # num_classes >>> criterion = LovaszHingeLoss() >>> pred = torch.randn(1, N, 3, 5, requires_grad=True) >>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(N) >>> output = criterion(pred, target) >>> output.backward() N)rcstdS)N)super__init__)self __class__rrr!szLovaszHingeLoss.__init__rcCs t||dS)N)rr)r)r"rrrrrforwardszLovaszHingeLoss.forward)__name__ __module__ __qualname____doc__r!rr% __classcell__rrr#rr[s,r) rZtorch.nnnnrZkornia.testingrrModulerrrrrs    P