o .wi@sddlmZddlZddlmZddlmZddlmZ ddeded ed ed d e eeeff d dZ dedeeefded efddZ  ddeded ed d ed ef ddZ dS))UnionN)Tensor)Literal)_mean_squared_error_updatemeanpredstarget num_outputs normalization)rrangestdl2returncCst|||\}}|dkr|dn|}|dkrtj|dd}n6|dkr4tj|ddjtj|ddj}n!|dkrAtj|ddd}n|d krNtj|d dd }nt d ||||fS) a[Updates and returns the sum of squared errors and the number of observations for NRMSE computation. Args: preds: Predicted tensor target: Ground truth tensor num_outputs: Number of outputs in multioutput setting normalization: type of normalization to be applied. Choose from "mean", "range", "std", "l2" rr)dimr r ) correctionrr )przQArgument `normalization` should be either 'mean', 'range', 'std' or 'l2' but got ) rviewtorchrmaxvaluesminr norm ValueError)rrr r sum_squared_errornum_obsdenomre/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/torchmetrics/functional/regression/nrmse.py*_normalized_root_mean_squared_error_updates " r!rrrcCst||}||S)z"Calculates RMSE and normalizes it.)rsqrt)rrrrmserrr +_normalized_root_mean_squared_error_compute5sr$rcCs"t||||d\}}}t|||S)afCalculates the `Normalized Root Mean Squared Error`_ (NRMSE) also know as scatter index. Args: preds: estimated labels target: ground truth labels normalization: type of normalization to be applied. Choose from "mean", "range", "std", "l2" which corresponds to normalizing the RMSE by the mean of the target, the range of the target, the standard deviation of the target or the L2 norm of the target. num_outputs: Number of outputs in multioutput setting Return: Tensor with the NRMSE score Example: >>> import torch >>> from torchmetrics.functional.regression import normalized_root_mean_squared_error >>> preds = torch.tensor([0., 1, 2, 3]) >>> target = torch.tensor([0., 1, 2, 2]) >>> normalized_root_mean_squared_error(preds, target, normalization="mean") tensor(0.4000) >>> normalized_root_mean_squared_error(preds, target, normalization="range") tensor(0.2500) >>> normalized_root_mean_squared_error(preds, target, normalization="std") tensor(0.6030) >>> normalized_root_mean_squared_error(preds, target, normalization="l2") tensor(0.1667) Example (multioutput): >>> import torch >>> from torchmetrics.functional.regression import normalized_root_mean_squared_error >>> preds = torch.tensor([[0., 1], [2, 3], [4, 5], [6, 7]]) >>> target = torch.tensor([[0., 1], [3, 3], [4, 5], [8, 9]]) >>> normalized_root_mean_squared_error(preds, target, normalization="mean", num_outputs=2) tensor([0.2981, 0.2222]) )r r )r!r$)rrr r rrrrrr "normalized_root_mean_squared_error=s* r%)r)rr) typingrrrtyping_extensionsr&torchmetrics.functional.regression.mserinttupler!r$r%rrrr sN