3jSSKJr SSKJr SSKrSSKJs Jr SSKJr SSK J r S S Sjjr "SS\R5r g) ) annotations)OptionalN)nn)mask_ignore_pixelsc B[U[R5(d[S[ U535e[ UR 5S:Xd[SUR 35eUR SSUR SS:Xd%[SUR SUR 35eURUR:Xd%[SURSUR35e[R"US S 9n[X5upURS URS 55nUb>URS 5RURS 9n X-nU R!S 5n O=UR upp[R""U 4X-URURS 9n UR!S 5n[R$"UX- [R&"XU-5SS9R)U5nUR+U5nSUR-5- $)uCCriterion that computes Tversky Coefficient loss. According to :cite:`salehi2017tversky`, we compute the Tversky Coefficient as follows: .. math:: \text{S}(P, G, \alpha; \beta) = \frac{|PG|}{|PG| + \alpha |P \setminus G| + \beta |G \setminus P|} Where: - :math:`P` and :math:`G` are the predicted and ground truth binary labels. - :math:`\alpha` and :math:`\beta` control the magnitude of the penalties for FPs and FNs, respectively. Note: - :math:`\alpha = \beta = 0.5` => dice coeff - :math:`\alpha = \beta = 1` => tanimoto coeff - :math:`\alpha + \beta = 1` => F beta coeff Args: pred: logits tensor with shape :math:`(N, C, H, W)` where C = number of classes. target: labels tensor with shape :math:`(N, H, W)` where each value is in range :math:`0 ≤ targets[i] ≤ C-1`. alpha: the first coefficient in the denominator. beta: the second coefficient in the denominator. eps: scalar for numerical stability. ignore_index: labels with this value are ignored in the loss computation. Return: the computed loss. Example: >>> N = 5 # num_classes >>> pred = torch.randn(1, N, 3, 5, requires_grad=True) >>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(N) >>> output = tversky_loss(pred, target, alpha=0.5, beta=0.5) >>> output.backward() z%pred type is not a torch.Tensor. Got z,Invalid pred shape, we expect BxNxHxW. Got: Nz.pred and target shapes must be the same. Got: z and z1pred and target must be in the same device. Got: )dim)dtype)r )r deviceg?)value) isinstancetorchTensor TypeErrortypelenshape ValueErrorrFsoftmaxrgather unsqueezetor sumfulladdcmul full_likeadd_divmean)predtargetalphabetaeps ignore_index pred_soft target_maskp_truemtotalB_HW intersection denominatorscores O/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/losses/tversky.py tversky_lossr8 s` dELL ) )?T |LMM tzz?a G |TUU ::bc?fll23/ /I$**UZ[a[g[gZhijj ;;&-- 'LT[[MY^_e_l_l^mnoo $A&I,VBF   a!1!1!!4 5F  ! !! $ ' 'djj ' 9i ZZ a A4djjM::i(L--  t|,   d3i    [ )E  c>^\rSrSrSrSSU4SjjjrSSjrSrU=r$) TverskyLossxuCriterion that computes Tversky Coefficient loss. According to :cite:`salehi2017tversky`, we compute the Tversky Coefficient as follows: .. math:: \text{S}(P, G, \alpha; \beta) = \frac{|PG|}{|PG| + \alpha |P \setminus G| + \beta |G \setminus P|} Where: - :math:`P` and :math:`G` are the predicted and ground truth binary labels. - :math:`\alpha` and :math:`\beta` control the magnitude of the penalties for FPs and FNs, respectively. Note: - :math:`\alpha = \beta = 0.5` => dice coeff - :math:`\alpha = \beta = 1` => tanimoto coeff - :math:`\alpha + \beta = 1` => F beta coeff Args: alpha: the first coefficient in the denominator. beta: the second coefficient in the denominator. eps: scalar for numerical stability. ignore_index: labels with this value are ignored in the loss computation. Shape: - Pred: :math:`(N, C, H, W)` where C = number of classes. - Target: :math:`(N, H, W)` where each value is :math:`0 ≤ targets[i] ≤ C-1`. Examples: >>> N = 5 # num_classes >>> criterion = TverskyLoss(alpha=0.5, beta=0.5) >>> pred = torch.randn(1, N, 3, 5, requires_grad=True) >>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(N) >>> output = criterion(pred, target) >>> output.backward() cR>[TU]5 XlX lX0lX@lgN)super__init__r'r(r)r*)selfr'r(r)r* __class__s r7r@TverskyLoss.__init__s# !  +7r9cp[XURURURUR5$r>)r8r'r(r)r*)rAr%r&s r7forwardTverskyLoss.forwards'D$**dii4K\K\]]r9)r'r(r)r*g:0yE>i) r'floatr(rHr)rHr* Optional[int]returnNone)r% torch.Tensorr&rLrJrL) __name__ __module__ __qualname____firstlineno____doc__r@rE__static_attributes__ __classcell__)rBs@r7r;r;xs'R88^^r9r;rG)r%rLr&rLr'rHr(rHr)rHr*rIrJrL) __future__rtypingrrtorch.nn.functionalr functionalrkornia.losses._utilsrr8Moduler;r9r7r[s$# 3"& U U U U  U  U  UUp2^"))2^r9