o oiw @spdZddlmZddlZddlmmZddlm Z dd d Z dd d Z dddZ ddddZ ddddZdS)zALosses based on the divergence between probability distributions.) annotationsN)TensorprqreturncCsX|j\}}}}tj||||||||||dd}|d||}|S)Nnone) reduction)shapeFkl_divreshapelogsumview)rrbatchchansheightwidth unsummed_kl kl_valuesrL/home/ubuntu/.local/lib/python3.10/site-packages/kornia/losses/divergence.py _kl_div_2ds *rcCs(d||}dt||dt||S)Ng?)r)rrmrrr _js_div_2d&s rlossesrstrcCs(|dkr|S|dkrt|St|S)Nrmean)torchrr)rrrrr _reduce_loss-sr rpredtargetcCtt|||S)aCalculate the Jensen-Shannon divergence loss between heatmaps. Args: pred: the input tensor with shape :math:`(B, N, H, W)`. target: the target 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 rr!r"rrrrjs_div_loss_2d3r%cCr#)aCalculate the Kullback-Leibler divergence loss between heatmaps. Args: pred: the input tensor with shape :math:`(B, N, H, W)`. target: the target 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 = kl_div_loss_2d(pred, pred) >>> loss.item() 0.0 )r rr$rrrkl_div_loss_2dIr&r')rrrrrr)rrrrrr)r)r!rr"rrrrr)__doc__ __future__rrtorch.nn.functionalnn functionalr kornia.corerrrr r%r'rrrrs