a  d @sddlmZddlZddlmZddlmZddlmZeeeeefdddZeee eeefd d d Z eeed d dZ eeedddZ dS))TupleN)Tensor) _check_data_shape_to_num_outputs)_check_same_shape)predstargetreturncCs&|jdkr||fS|d|dfS)N)ndimZ unsqueeze)rrr t/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/torchmetrics/functional/regression/log_cosh.py_unsqueeze_tensorss r)rr num_outputsrcCstt||t|||t||\}}||}tt|t| dd}tj|j d|j d}||fS)aTUpdate and returns variables required to compute LogCosh error. Check for same shape of input tensors. Args: preds: Predicted tensor target: Ground truth tensor num_outputs: Number of outputs in multioutput setting Return: Sum of LogCosh error over examples, and total number of examples r r)device) rrrtorchlogexpsumsqueezeZtensorshaper)rrrdiffsum_log_cosh_errorn_obsr r r _log_cosh_error_updates  *r)rrrcCs ||S)zCompute Mean Squared Error. Args: sum_log_cosh_error: Sum of LogCosh errors over all observations n_obs: Number of predictions or observations )r)rrr r r _log_cosh_error_compute5srcCs0t|||jdkrdn|jdd\}}t||S)aZCompute the `LogCosh Error`_. .. math:: \text{LogCoshError} = \log\left(\frac{\exp(\hat{y} - y) + \exp(\hat{y - y})}{2}\right) Where :math:`y` is a tensor of target values, and :math:`\hat{y}` is a tensor of predictions. Args: preds: estimated labels with shape ``(batch_size,)`` or `(batch_size, num_outputs)`` target: ground truth labels with shape ``(batch_size,)`` or `(batch_size, num_outputs)`` Return: Tensor with LogCosh error Example (single output regression):: >>> from torchmetrics.functional.regression import log_cosh_error >>> preds = torch.tensor([3.0, 5.0, 2.5, 7.0]) >>> target = torch.tensor([2.5, 5.0, 4.0, 8.0]) >>> log_cosh_error(preds, target) tensor(0.3523) Example (multi output regression):: >>> from torchmetrics.functional.regression import log_cosh_error >>> preds = torch.tensor([[3.0, 5.0, 1.2], [-2.1, 2.5, 7.0]]) >>> target = torch.tensor([[2.5, 5.0, 1.3], [0.3, 4.0, 8.0]]) >>> log_cosh_error(preds, target) tensor([0.9176, 0.4277, 0.2194]) r )r)rr rr)rrrrr r r log_cosh_error@s r) typingrrrZ(torchmetrics.functional.regression.utilsrZtorchmetrics.utilities.checksrrintrrrr r r r s