i ddlZddlmZddlZddlmZddlmZddlmZddl m Z m Z m Z m Z mZmZ ddeded ed ed d eed ef dZdeded efdZ ddededed ed d eed ef dZ ddeded ed d eed ef dZy)N)Optional)Tensor)Literal)#_multiclass_confusion_matrix_update)_compute_bias_corrected_values_compute_chi_squared_drop_empty_rows_and_cols_handle_nan_in_data_nominal_input_validation&_unable_to_use_bias_correction_warningpredstarget num_classes nan_strategy)replacedropnan_replace_valuereturnc|jdk(r|jdn|}|jdk(r|jdn|}t||||\}}t|||S)aCompute the bins to update the confusion matrix with for Cramer's V calculation. Args: preds: 1D or 2D tensor of categorical (nominal) data target: 1D or 2D tensor of categorical (nominal) data num_classes: Integer specifying the number of classes nan_strategy: Indication of whether to replace or drop ``NaN`` values nan_replace_value: Value to replace ``NaN`s when ``nan_strategy = 'replace``` Returns: Non-reduced confusion matrix )ndimargmaxr r)r rrrrs |/Volumes/fast/ai/experiments/voice-extract-mac/.venv/lib/python3.12/site-packages/torchmetrics/functional/nominal/cramers.py_cramers_v_updater s\( %zzQELLOEE!'!1V]]1 vF'v|EVWME6 .ufk JJconfmatbias_correctionct|}|j}t||}||z }|j\}}|rt ||||\}}} t j || dk(r6tdt jtd|jSt j|t j |dz | dz z } n(t j|t |dz |dz z } | jddS)zCompute Cramers' V statistic based on a pre-computed confusion matrix. Args: confmat: Confusion matrix for observed data bias_correction: Indication of whether to use bias correction. Returns: Cramer's V statistic rz Cramer's V) metric_namenandeviceg?) r sumrshapertorchminr tensorfloatr#sqrtclamp) rrcm_sum chi_squared phi_squarednum_rowsnum_colsphi_squared_correctedrows_correctedcols_correctedcramers_v_values r_cramers_v_computer6:s(0G [[]F&w@K&K Hh@^ 8VA =~~ 99^^ 4 9 2| L<<e W^^D D**%:UYY~XYGY[ilm[m=n%no**[3x!|XPQ\3R%RS  c **rct||ttj||gj }t |||||}t ||S)a Compute `Cramer's V`_ statistic measuring the association between two categorical (nominal) data series. .. math:: V = \sqrt{\frac{\chi^2 / n}{\min(r - 1, k - 1)}} where .. math:: \chi^2 = \sum_{i,j} \ frac{\left(n_{ij} - \frac{n_{i.} n_{.j}}{n}\right)^2}{\frac{n_{i.} n_{.j}}{n}} where :math:`n_{ij}` denotes the number of times the values :math:`(A_i, B_j)` are observed with :math:`A_i, B_j` represent frequencies of values in ``preds`` and ``target``, respectively. Cramer's V is a symmetric coefficient, i.e. :math:`V(preds, target) = V(target, preds)`. The output values lies in [0, 1] with 1 meaning the perfect association. Args: preds: 1D or 2D tensor of categorical (nominal) data - 1D shape: (batch_size,) - 2D shape: (batch_size, num_classes) target: 1D or 2D tensor of categorical (nominal) data - 1D shape: (batch_size,) - 2D shape: (batch_size, num_classes) bias_correction: Indication of whether to use bias correction. nan_strategy: Indication of whether to replace or drop ``NaN`` values nan_replace_value: Value to replace ``NaN``s when ``nan_strategy = 'replace'`` Returns: Cramer's V statistic Example: >>> from torch import randint, round >>> from torchmetrics.functional.nominal import cramers_v >>> preds = randint(0, 4, (100,)) >>> target = round(preds + torch.randn(100)).clamp(0, 4) >>> cramers_v(preds, target) tensor(0.5284) )r lenr'catuniquerr6)r rrrrrrs r cramers_vr;XsP^l,=>eii0779:Kv{LJ[\G g 77rmatrixct|||jd}tj|||j}t j t|dD]l\}}|dd|f|dd|f} }ttj|| gj} t|| | ||} t| |x|||f<|||f<n|S)aCompute `Cramer's V`_ statistic between a set of multiple variables. This can serve as a convenient tool to compute Cramer's V statistic for analyses of correlation between categorical variables in your dataset. Args: matrix: A tensor of categorical (nominal) data, where: - rows represent a number of data points - columns represent a number of categorical (nominal) features bias_correction: Indication of whether to use bias correction. nan_strategy: Indication of whether to replace or drop ``NaN`` values nan_replace_value: Value to replace ``NaN``s when ``nan_strategy = 'replace'`` Returns: Cramer's V statistic for a dataset of categorical variables Example: >>> from torch import randint >>> from torchmetrics.functional.nominal import cramers_v_matrix >>> matrix = randint(0, 4, (200, 5)) >>> cramers_v_matrix(matrix) tensor([[1.0000, 0.0637, 0.0000, 0.0542, 0.1337], [0.0637, 1.0000, 0.0000, 0.0000, 0.0000], [0.0000, 0.0000, 1.0000, 0.0000, 0.0649], [0.0542, 0.0000, 0.0000, 1.0000, 0.1100], [0.1337, 0.0000, 0.0649, 0.1100, 1.0000]]) rr"rN) r r&r'onesr# itertools combinationsranger8r9r:rr6) r<rrr num_variablescramers_v_matrix_valueijxyrrs rcramers_v_matrixrHsDl,=>LLOM"ZZ }V]][&&u]';Q?1ad|VAqD\1%))QF+2245 #Aq+|EVWFXY`bqFrrq!t$'=ad'C @ "!r)rr$)Trr$)r?typingrr'rtyping_extensionsr7torchmetrics.functional.classification.confusion_matrixr%torchmetrics.functional.nominal.utilsrrr r r r intr*rboolr6r;rHrrrPs/ %g09), K K KK+, K   K  K4+++&+B!/8), 28 28 2828+, 28   28  28n!/8), *" *"*"+,*"  *"  *"r