o siu @sFddlmZGdddeZGdddeZeZeZeZeZdS))_Lossc@eZdZdZddZdS) PairwiseMSEa6Measure pairwise mean square error on a batch. Shape: - est_targets : :math:`(batch, nsrc, ...)`. - targets: :math:`(batch, nsrc, ...)`. Returns: :class:`torch.Tensor`: with shape :math:`(batch, nsrc, nsrc)` Examples >>> import torch >>> from asteroid.losses import PITLossWrapper >>> targets = torch.randn(10, 2, 32000) >>> est_targets = torch.randn(10, 2, 32000) >>> loss_func = PITLossWrapper(PairwiseMSE(), pit_from='pairwise') >>> loss = loss_func(est_targets, targets) cCst||ks |jdkrtd|d|d|d}|d}||d}ttd|j}|j|dS)Nz/Inputs must be of shape [batch, n_src, *], got  and  insteaddim)sizendim TypeError unsqueezelistrangemean)self est_targetstargetspw_loss mean_overrG/home/ubuntu/.local/lib/python3.10/site-packages/asteroid/losses/mse.pyforwards    zPairwiseMSE.forwardN__name__ __module__ __qualname____doc__rrrrrrs rc@r) SingleSrcMSEaMeasure mean square error on a batch. Supports both tensors with and without source axis. Shape: - est_targets: :math:`(batch, ...)`. - targets: :math:`(batch, ...)`. Returns: :class:`torch.Tensor`: with shape :math:`(batch)` Examples >>> import torch >>> from asteroid.losses import PITLossWrapper >>> targets = torch.randn(10, 2, 32000) >>> est_targets = torch.randn(10, 2, 32000) >>> # singlesrc_mse / multisrc_mse support both 'pw_pt' and 'perm_avg'. >>> loss_func = PITLossWrapper(singlesrc_mse, pit_from='pw_pt') >>> loss = loss_func(est_targets, targets) cCs`||ks |jdkrtd|d|d||d}ttd|j}|j|dS)Nr z(Inputs must be of shape [batch, *], got rrrr )r r rrrr)rrrlossrrrrr9s  zSingleSrcMSE.forwardNrrrrrr $s r N)torch.nn.modules.lossrrr MultiSrcMSE pairwise_mse singlesrc_mse multisrc_mserrrrs