o yiD*@sddlmZmZddlZddlmZddlmZmZmZm Z m Z m Z m Z ddl mZGdddeZGdd d eZGd d d eZdS) )AnyOptionalN)Tensor)+_multilabel_confusion_matrix_arg_validation#_multilabel_confusion_matrix_format!_multilabel_coverage_error_update,_multilabel_ranking_average_precision_update_multilabel_ranking_loss_update%_multilabel_ranking_tensor_validation_ranking_reduce)Metricc eZdZUdZdZeed<dZeed<dZeed<  dde d e e d ed e d df fd d Z de de d dfddZd e fddZZS)MultilabelCoverageErroraComputes `Multilabel coverage error`_. The score measure how far we need to go through the ranked scores to cover all true labels. The best value is equal to the average number of labels in the target tensor per sample. As input to ``forward`` and ``update`` the metric accepts the following input: - ``preds`` (:class:`~torch.Tensor`): A float tensor of shape ``(N, C, ...)``. Preds should be a tensor containing probabilities or logits for each observation. If preds has values outside [0,1] range we consider the input to be logits and will auto apply sigmoid per element. - ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, C, ...)``. Target should be a tensor containing ground truth labels, and therefore only contain {0,1} values (except if `ignore_index` is specified). .. note:: Additional dimension ``...`` will be flattened into the batch dimension. As output to ``forward`` and ``compute`` the metric returns the following output: - ``mlce`` (:class:`~torch.Tensor`): A tensor containing the multilabel coverage error. Args: num_labels: Integer specifing the number of labels ignore_index: Specifies a target value that is ignored and does not contribute to the metric calculation validate_args: bool indicating if input arguments and tensors should be validated for correctness. Set to ``False`` for faster computations. Example: >>> from torchmetrics.classification import MultilabelCoverageError >>> _ = torch.manual_seed(42) >>> preds = torch.rand(10, 5) >>> target = torch.randint(2, (10, 5)) >>> mlce = MultilabelCoverageError(num_labels=5) >>> mlce(preds, target) tensor(3.9000) Fhigher_is_betteris_differentiablefull_state_updateNT num_labels ignore_index validate_argskwargsreturnc ftjdi||rt|d|d||_||_||_|jdtddd|jdtddddSN) thresholdrmeasuresum)dist_reduce_fxtotal super__init__rrrr add_statetorchtensorselfrrrr __class__rW/home/ubuntu/.local/lib/python3.10/site-packages/torchmetrics/classification/ranking.pyr"Gz MultilabelCoverageError.__init__predstargetcCb|jr t|||j|jt|||jd|jdd\}}t||\}}|j|7_|j|7_dSNrF)rrshould_threshold)rr rrrrrrr'r,r-r n_elementsrrr*updateW zMultilabelCoverageError.updatecCt|j|jSNr rrr'rrr*computeazMultilabelCoverageError.computeNT__name__ __module__ __qualname____doc__rbool__annotations__rrintrrr"rr3r9 __classcell__rrr(r*rs(  #   rc seZdZUdZdZeed<dZeed<dZeed<  dde d e e d ed e d df fd d Z de de d dfddZd e fddZZS)!MultilabelRankingAveragePrecisiona'Computes label ranking average precision score for multilabel data [1]. The score is the average over each ground truth label assigned to each sample of the ratio of true vs. total labels with lower score. Best score is 1. As input to ``forward`` and ``update`` the metric accepts the following input: - ``preds`` (:class:`~torch.Tensor`): A float tensor of shape ``(N, C, ...)``. Preds should be a tensor containing probabilities or logits for each observation. If preds has values outside [0,1] range we consider the input to be logits and will auto apply sigmoid per element. - ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, C, ...)``. Target should be a tensor containing ground truth labels, and therefore only contain {0,1} values (except if `ignore_index` is specified). .. note:: Additional dimension ``...`` will be flattened into the batch dimension. As output to ``forward`` and ``compute`` the metric returns the following output: - ``mlrap`` (:class:`~torch.Tensor`): A tensor containing the multilabel ranking average precision. Args: num_labels: Integer specifing the number of labels ignore_index: Specifies a target value that is ignored and does not contribute to the metric calculation validate_args: bool indicating if input arguments and tensors should be validated for correctness. Set to ``False`` for faster computations. Example: >>> from torchmetrics.classification import MultilabelRankingAveragePrecision >>> _ = torch.manual_seed(42) >>> preds = torch.rand(10, 5) >>> target = torch.randint(2, (10, 5)) >>> mlrap = MultilabelRankingAveragePrecision(num_labels=5) >>> mlrap(preds, target) tensor(0.7744) TrFrrNrrrrrc rrr r&r(rr*r"r+z*MultilabelRankingAveragePrecision.__init__r,r-cCr.r/)rr rrrrrrr1rrr*r3r4z(MultilabelRankingAveragePrecision.updatecCr5r6r7r8rrr*r9r:z)MultilabelRankingAveragePrecision.computer;r<rrr(r*rEes(  $   rEc r )MultilabelRankingLossaComputes the label ranking loss for multilabel data [1]. The score is corresponds to the average number of label pairs that are incorrectly ordered given some predictions weighted by the size of the label set and the number of labels not in the label set. The best score is 0. As input to ``forward`` and ``update`` the metric accepts the following input: - ``preds`` (:class:`~torch.Tensor`): A float tensor of shape ``(N, C, ...)``. Preds should be a tensor containing probabilities or logits for each observation. If preds has values outside [0,1] range we consider the input to be logits and will auto apply sigmoid per element. - ``target`` (:class:`~torch.Tensor`): An int tensor of shape ``(N, C, ...)``. Target should be a tensor containing ground truth labels, and therefore only contain {0,1} values (except if `ignore_index` is specified). .. note:: Additional dimension ``...`` will be flattened into the batch dimension. As output to ``forward`` and ``compute`` the metric returns the following output: - ``mlrl`` (:class:`~torch.Tensor`): A tensor containing the multilabel ranking loss. Args: preds: Tensor with predictions target: Tensor with true labels num_labels: Integer specifing the number of labels ignore_index: Specifies a target value that is ignored and does not contribute to the metric calculation validate_args: bool indicating if input arguments and tensors should be validated for correctness. Set to ``False`` for faster computations. Example: >>> from torchmetrics.classification import MultilabelRankingLoss >>> _ = torch.manual_seed(42) >>> preds = torch.rand(10, 5) >>> target = torch.randint(2, (10, 5)) >>> mlrl = MultilabelRankingLoss(num_labels=5) >>> mlrl(preds, target) tensor(0.4167) FrrrNTrrrrrc rrr r&r(rr*r"r+zMultilabelRankingLoss.__init__r,r-cCr.r/)rr rrrr rrr1rrr*r3r4zMultilabelRankingLoss.updatecCr5r6r7r8rrr*r9r:zMultilabelRankingLoss.computer;r<rrr(r*rFs(  &   rF)typingrrr$r.torchmetrics.functional.classification.rankingrrrrr r r torchmetrics.metricr rrErFrrrr*s  $ FG