a  d@sddlmZmZddlZddlmZddlmZddlmZddl m Z eeeeefddd Z deeee e fed ed ddZdeeee e fed ed ddZdS))TupleUnionN)Tensor)Literal)_check_same_shape)reduce)predstargetreturncCs`|j|jkr&td|jd|jdt||t|jdkrXtd|jd|jd||fS)uUpdate and returns variables required to compute Erreur Relative Globale Adimensionnelle de Synthèse. Args: preds: Predicted tensor target: Ground truth tensor zEExpected `preds` and `target` to have the same data type. Got preds: z and target: .z@Expected `preds` and `target` to have BxCxHxW shape. Got preds: )Zdtype TypeErrorrlenshape ValueError)rr rl/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/torchmetrics/functional/image/ergas.py _ergas_updates$  rr elementwise_mean)rsumnoneN)rr ratio reductionr c Cs|j\}}}}|||||}|||||}||}tj||dd} t| ||} tj|dd} d|ttj| | ddd|} t| |S)uErreur Relative Globale Adimensionnelle de Synthèse. Args: preds: estimated image target: ground truth image ratio: ratio of high resolution to low resolution reduction: a method to reduce metric score over labels. - ``'elementwise_mean'``: takes the mean (default) - ``'sum'``: takes the sum - ``'none'`` or ``None``: no reduction will be applied Example: >>> gen = torch.manual_seed(42) >>> preds = torch.rand([16, 1, 16, 16], generator=gen) >>> target = preds * 0.75 >>> preds, target = _ergas_update(preds, target) >>> torch.round(_ergas_compute(preds, target)) tensor(154.) )Zdimd)rZreshapetorchrsqrtmeanr) rr rrbchwdiffZsum_squared_errorZ rmse_per_bandZ mean_targetZ ergas_scorerrr_ergas_compute.s(r$cCst||\}}t||||S)u;Erreur Relative Globale Adimensionnelle de Synthèse. Args: preds: estimated image target: ground truth image ratio: ratio of high resolution to low resolution reduction: a method to reduce metric score over labels. - ``'elementwise_mean'``: takes the mean (default) - ``'sum'``: takes the sum - ``'none'`` or ``None``: no reduction will be applied Return: Tensor with RelativeG score Raises: TypeError: If ``preds`` and ``target`` don't have the same data type. ValueError: If ``preds`` and ``target`` don't have ``BxCxHxW shape``. Example: >>> from torchmetrics.functional.image import error_relative_global_dimensionless_synthesis >>> gen = torch.manual_seed(42) >>> preds = torch.rand([16, 1, 16, 16], generator=gen) >>> target = preds * 0.75 >>> ergds = error_relative_global_dimensionless_synthesis(preds, target) >>> torch.round(ergds) tensor(154.) References: [1] Qian Du; Nicholas H. Younan; Roger King; Vijay P. Shah, "On the Performance Evaluation of Pan-Sharpening Techniques" in IEEE Geoscience and Remote Sensing Letters, vol. 4, no. 4, pp. 518-522, 15 October 2007, doi: 10.1109/LGRS.2007.896328. )rr$)rr rrrrr-error_relative_global_dimensionless_synthesisVs*r%)r r)r r)typingrrrrZtyping_extensionsrZtorchmetrics.utilities.checksrZ"torchmetrics.utilities.distributedrrintfloatr$r%rrrrs0      +