iSCDddlmZddlZddlmZddlmZddlmZmZm Z m Z m Z m Z m Z mZmZmZmZmZddlmZddlmZ dded eed d eed ed ef dZ ddededed eeded ed efdZ d ded eed eedd dfdZ d!dededed eedd eeded ed efdZ d"deded eed eedd df dZ d#dedededed eedd eeded ed efdZ d$dedededdedeedeed eedd eeded ed efdZ y)%)OptionalN)Tensor)Literal) '_binary_confusion_matrix_arg_validation_binary_confusion_matrix_format*_binary_confusion_matrix_tensor_validation_binary_confusion_matrix_update+_multiclass_confusion_matrix_arg_validation#_multiclass_confusion_matrix_format._multiclass_confusion_matrix_tensor_validation#_multiclass_confusion_matrix_update+_multilabel_confusion_matrix_arg_validation#_multilabel_confusion_matrix_format._multilabel_confusion_matrix_tensor_validation#_multilabel_confusion_matrix_update) _safe_divide)ClassificationTaskconfmataverage)micromacroweightednonebinary ignore_index zero_divisionreturncgd}||vrtd|d|d|j}|dk(r t|d|d|dz|dz| S|d uxrd |cxkxr|jd knc}|jd k(}|r+|d d d d f}|d d d d f|d d d d fz|d d d d fz}n;t j |}|jd |jd z|z }|dk(r*|j}|j|r||ndz }t||| } | |dk(s|dk(r| S|dk(r6|jd k(r|d d d d f|d d d d fzn|jd } nGt j| } |rd| |<|s)d| |jd |jd zd k(<| | z| jz jS)a'Perform reduction of an un-normalized confusion matrix into jaccard score. Args: confmat: tensor with un-normalized confusionmatrix average: reduction method - ``'binary'``: binary reduction, expects a 2x2 matrix - ``'macro'``: Calculate the metric for each class separately, and average the metrics across classes (with equal weights for each class). - ``'micro'``: Calculate the metric globally, across all samples and classes. - ``'weighted'``: Calculate the metric for each class separately, and average the metrics across classes, weighting each class by its support (``tp + fn``). - ``'none'`` or ``None``: Calculate the metric for each class separately, and return the metric for every class. ignore_index: Specifies a target value that is ignored and does not contribute to the metric calculation zero_division: Value to replace when there is a division by zero. Should be `0` or `1`. )rrrrrNzThe `average` has to be one of z, got .r)r )rr )r r)rNrr rrr) ValueErrorfloatrshapendimtorchdiagsum ones_like) rrrrallowed_averageignore_index_cond multilabelnumdenomjaccardweightss /Volumes/fast/ai/experiments/voice-extract-mac/.venv/lib/python3.12/site-packages/torchmetrics/functional/classification/jaccard.py_jaccard_index_reducer3&s6MOo%:?:K6RYQZZ[\]]mmoG(GDMGDMGDM,IGTXM,Yjwxx$D0YQ,5YWXIY5Y"JaAg1a 71a7#33gaAg6FFjj! AQ/#5'ggi 6Gu\2SQ3]CG'V+w'/A*9@9J'!Q'"WQ1W%55PWP[P[\]P^//'* $'GL !>> from torch import tensor >>> from torchmetrics.functional.classification import binary_jaccard_index >>> target = tensor([1, 1, 0, 0]) >>> preds = tensor([0, 1, 0, 0]) >>> binary_jaccard_index(preds, target) tensor(0.5000) Example (preds is float tensor): >>> from torchmetrics.functional.classification import binary_jaccard_index >>> target = tensor([1, 1, 0, 0]) >>> preds = tensor([0.35, 0.85, 0.48, 0.01]) >>> binary_jaccard_index(preds, target) tensor(0.5000) r)rr)rrrr r3)r5r6r7rr8rrs r2binary_jaccard_indexr:dsMh/ <H25&,O3E69l[ME6-eVs@ 0 \JBOo%D_DUU_`g_hhijkk&r4c|rt|||t||||t|||\}}t|||}t ||||S)a" Calculate the Jaccard index for multiclass tasks. The `Jaccard index`_ (also known as the intersection over union or jaccard similarity coefficient) is an statistic that can be used to determine the similarity and diversity of a sample set. It is defined as the size of the intersection divided by the union of the sample sets: .. math:: J(A,B) = \frac{|A\cap B|}{|A\cup B|} Accepts the following input tensors: - ``preds``: ``(N, ...)`` (int tensor) or ``(N, C, ..)`` (float tensor). If preds is a floating point we apply ``torch.argmax`` along the ``C`` dimension to automatically convert probabilities/logits into an int tensor. - ``target`` (int tensor): ``(N, ...)`` Additional dimension ``...`` will be flattened into the batch dimension. Args: preds: Tensor with predictions target: Tensor with true labels num_classes: Integer specifying the number of classes average: Defines the reduction that is applied over labels. Should be one of the following: - ``micro``: Sum statistics over all labels - ``macro``: Calculate statistics for each label and average them - ``weighted``: calculates statistics for each label and computes weighted average using their support - ``"none"`` or ``None``: calculates statistic for each label and applies no reduction 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. zero_division: Value to replace when there is a division by zero. Should be `0` or `1`. Example (pred is integer tensor): >>> from torch import tensor >>> from torchmetrics.functional.classification import multiclass_jaccard_index >>> target = tensor([2, 1, 0, 0]) >>> preds = tensor([2, 1, 0, 1]) >>> multiclass_jaccard_index(preds, target, num_classes=3) tensor(0.6667) Example (pred is float tensor): >>> from torchmetrics.functional.classification import multiclass_jaccard_index >>> target = tensor([2, 1, 0, 0]) >>> preds = tensor([[0.16, 0.26, 0.58], ... [0.22, 0.61, 0.17], ... [0.71, 0.09, 0.20], ... [0.05, 0.82, 0.13]]) >>> multiclass_jaccard_index(preds, target, num_classes=3) tensor(0.6667) rrr)r>r r r r3)r5r6r;rrr8rrs r2multiclass_jaccard_indexrAsU@0lGT6ufkS_`7v|TME61%MG ' dq rrr4 num_labelscNt|||d}||vrtd|d|dyr=)rr#)rBr7rrr+s r2(_multilabel_jaccard_index_arg_validationrDsB 0 I|TBOo%D_DUU_`g_hhijkk&r4c|rt|||t||||t|||||\}}t|||}t ||||S)aZ Calculate the Jaccard index for multilabel tasks. The `Jaccard index`_ (also known as the intersection over union or jaccard similarity coefficient) is an statistic that can be used to determine the similarity and diversity of a sample set. It is defined as the size of the intersection divided by the union of the sample sets: .. math:: J(A,B) = \frac{|A\cap B|}{|A\cup B|} Accepts the following input tensors: - ``preds`` (int or float tensor): ``(N, C, ...)``. If preds is a floating point tensor with values outside [0,1] range we consider the input to be logits and will auto apply sigmoid per element. Additionally, we convert to int tensor with thresholding using the value in ``threshold``. - ``target`` (int tensor): ``(N, C, ...)`` Additional dimension ``...`` will be flattened into the batch dimension. Args: preds: Tensor with predictions target: Tensor with true labels num_labels: Integer specifying the number of labels threshold: Threshold for transforming probability to binary (0,1) predictions average: Defines the reduction that is applied over labels. Should be one of the following: - ``micro``: Sum statistics over all labels - ``macro``: Calculate statistics for each label and average them - ``weighted``: calculates statistics for each label and computes weighted average using their support - ``"none"`` or ``None``: calculates statistic for each label and applies no reduction 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. zero_division: Value to replace when there is a division by zero. Should be `0` or `1`. Example (preds is int tensor): >>> from torch import tensor >>> from torchmetrics.functional.classification import multilabel_jaccard_index >>> target = tensor([[0, 1, 0], [1, 0, 1]]) >>> preds = tensor([[0, 0, 1], [1, 0, 1]]) >>> multilabel_jaccard_index(preds, target, num_labels=3) tensor(0.5000) Example (preds is float tensor): >>> from torchmetrics.functional.classification import multilabel_jaccard_index >>> target = tensor([[0, 1, 0], [1, 0, 1]]) >>> preds = tensor([[0.11, 0.22, 0.84], [0.73, 0.33, 0.92]]) >>> multilabel_jaccard_index(preds, target, num_labels=3) tensor(0.5000) r@)rDrrrr3) r5r6rBr7rrr8rrs r2multilabel_jaccard_indexrFs[~0Y U6ufjR^_7vzS\^jkME61%LG ' dq rrr4task)r multiclassr-c tj|}|tjk(rt|||||| S|tjk(r9t |t stdt|dt||||||| S|tjk(r:t |t stdt|dt|||||||| Std|)aCalculate the Jaccard index. The `Jaccard index`_ (also known as the intersection over union or jaccard similarity coefficient) is an statistic that can be used to determine the similarity and diversity of a sample set. It is defined as the size of the intersection divided by the union of the sample sets: .. math:: J(A,B) = \frac{|A\cap B|}{|A\cup B|} This function is a simple wrapper to get the task specific versions of this metric, which is done by setting the ``task`` argument to either ``'binary'``, ``'multiclass'`` or ``'multilabel'``. See the documentation of :func:`~torchmetrics.functional.classification.binary_jaccard_index`, :func:`~torchmetrics.functional.classification.multiclass_jaccard_index` and :func:`~torchmetrics.functional.classification.multilabel_jaccard_index` for the specific details of each argument influence and examples. Legacy Example: >>> from torch import randint, tensor >>> target = randint(0, 2, (10, 25, 25)) >>> pred = tensor(target) >>> pred[2:5, 7:13, 9:15] = 1 - pred[2:5, 7:13, 9:15] >>> jaccard_index(pred, target, task="multiclass", num_classes=2) tensor(0.9660) z+`num_classes` is expected to be `int` but `z was passed.`z*`num_labels` is expected to be `int` but `zNot handled value: ) rfrom_strBINARYr: MULTICLASS isinstanceintr#typerA MULTILABELrF) r5r6rGr7r;rBrrr8rs r2 jaccard_indexrQFsH  & &t ,D !(((#E69lM[hii !,,,+s+J4P[K\J]]jkl l'v{G\[hjwxx !,,,*c*I$zJZI[[hij j' 6:y'<Xe   *4&1 22r4)Nr")?NTr")NN)rNTr")rRNr)rRrNTr")rRNNrNTr")!typingrr'rtyping_extensionsr7torchmetrics.functional.classification.confusion_matrixrrrr r r r r rrrrtorchmetrics.utilities.computertorchmetrics.utilities.enumsrrNr$r3boolr:r>rArDrFrQr4r2rZs* %    8; #' ;7 ;7 gLM N;73-;7 ;7  ;7B"& 9Y 9Y 9Y9Y3- 9Y  9Y  9Y 9Y|#'GKll3-lgBC Dl lHO"&Es Es EsEsgBC D Es 3- Es  EsEs EsT"&GN l l l3- lgBC D l  l GN"&Ds Ds DsDs Ds gBC D Ds 3- DsDsDs DsV!% $GN"&13 13 13 6 713 13 # 13  13gBC D133-131313 13r4