a dk@shddlZddlmZddlmmZddlmZd ejeje e e ejdddZ Gdddej Z dS) N)one_hot:0yE>)inputtargetalphabetaepsreturncCs0t|tjstdt|t|jdks dice coeff - :math:`\alpha = \beta = 1` => tanimoto coeff - :math:`\alpha + \beta = 1` => F beta coeff Args: input: logits tensor with shape :math:`(N, C, H, W)` where C = number of classes. target: labels tensor with shape :math:`(N, H, W)` where each value is :math:`0 ≤ targets[i] ≤ C−1`. alpha: the first coefficient in the denominator. beta: the second coefficient in the denominator. eps: scalar for numerical stability. Return: the computed loss. Example: >>> N = 5 # num_classes >>> input = torch.randn(1, N, 3, 5, requires_grad=True) >>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(N) >>> output = tversky_loss(input, target, alpha=0.5, beta=0.5) >>> output.backward() z&Input type is not a torch.Tensor. Got z-Invalid input shape, we expect BxNxHxW. Got: Nz/input and target shapes must be the same. Got: z and z2input and target must be in the same device. Got: )Zdim)Z num_classesdevicedtype)r g?) isinstancetorchTensor TypeErrortypelenshape ValueErrorr FZsoftmaxrrsummean)rrrrrZ input_softZtarget_one_hotdims intersectionZfpsfns numerator denominator tversky_lossr"^/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/kornia/losses/tversky.pyr! s$)   r!csFeZdZdZd eeeddfdd Zejejejddd ZZ S) TverskyLossu<Criterion that computes Tversky Coefficient loss. According to :cite:`salehi2017tversky`, we compute the Tversky Coefficient as follows: .. math:: \text{S}(P, G, \alpha; \beta) = \frac{|PG|}{|PG| + \alpha |P \setminus G| + \beta |G \setminus P|} Where: - :math:`P` and :math:`G` are the predicted and ground truth binary labels. - :math:`\alpha` and :math:`\beta` control the magnitude of the penalties for FPs and FNs, respectively. Note: - :math:`\alpha = \beta = 0.5` => dice coeff - :math:`\alpha = \beta = 1` => tanimoto coeff - :math:`\alpha + \beta = 1` => F beta coeff Args: alpha: the first coefficient in the denominator. beta: the second coefficient in the denominator. eps: scalar for numerical stability. Shape: - Input: :math:`(N, C, H, W)` where C = number of classes. - Target: :math:`(N, H, W)` where each value is :math:`0 ≤ targets[i] ≤ C−1`. Examples: >>> N = 5 # num_classes >>> criterion = TverskyLoss(alpha=0.5, beta=0.5) >>> input = torch.randn(1, N, 3, 5, requires_grad=True) >>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(N) >>> output = criterion(input, target) >>> output.backward() rN)rrrr cs t||_||_||_dSN)super__init__rrr)selfrrr __class__r"r#r'{s zTverskyLoss.__init__)rrr cCst|||j|j|jSr%)r!rrr)r(rrr"r"r#forwardszTverskyLoss.forward)r) __name__ __module__ __qualname____doc__floatr'rrr+ __classcell__r"r"r)r#r$Ss'r$)r) rZtorch.nnnnZtorch.nn.functionalZ functionalrZkornia.utils.one_hotrrr0r!Moduler$r"r"r"r#s   H