3j hSrSSKJr SSKrSSKJs Jr S SjrS Sjr S Sjr S S Sjjr S S Sjjr g) zALosses based on the divergence between probability distributions.) annotationsNcURup#pE[R"URX#-XE-5R 5URX#-XE-5SS9nUR S5R X#5nU$)Nnone) reduction)shapeFkl_divreshapelogsumview)pqbatchchansheightwidth unsummed_kl kl_valuess R/home/wildlama/miniconda3/lib/python3.13/site-packages/kornia/losses/divergence.py _kl_div_2drsp"#''E&(( %-0446 %-QWQ_8`lrK#((6I cJSX--nS[X5-S[X5--$)Ng?)r)rrms r _js_div_2dr$s, qu A A! !C*Q*:$: ::rcvUS:XaU$US:Xa[R"U5$[R"U5$)Nrmean)torchrr )lossesrs r _reduce_lossr!+s3F !*f!45::f K%))F:KKrc,[[X5U5$)auCalculate the Jensen-Shannon divergence loss between heatmaps. Args: pred: the input torch.Tensor with shape :math:`(B, N, H, W)`. target: the target torch.Tensor with shape :math:`(B, N, H, W)`. 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. Examples: >>> pred = torch.full((1, 1, 2, 4), 0.125) >>> loss = js_div_loss_2d(pred, pred) >>> loss.item() 0.0 )r!rpredtargetrs rjs_div_loss_2dr&1&  60) <>> pred = torch.full((1, 1, 2, 4), 0.125) >>> loss = kl_div_loss_2d(pred, pred) >>> loss.item() 0.0 )r!rr#s rkl_div_loss_2dr)Gr'r)r torch.Tensorrr*returnr*)r r*rstrr+r*)r)r$r*r%r*rr,r+r*) __doc__ __future__rrtorch.nn.functionalnn functionalr rrr!r&r)rrr3s1$I" ;L =,=r