o .wi, @sddlmZddlZddlmZddlmZdededeeeffdd Zddeded eedefd d Z ddeded eedefddZ dS))OptionalN)Tensor)_check_same_shapepredstargetreturncCs<t|||jdkrtd|j|}|}||fS)zUpdate and returns variables required to compute Cosine Similarity. Checks for same shape of input tensors. Args: preds: Predicted tensor target: Ground truth tensor zExpected input to cosine similarity to be 2D tensors of shape `[N,D]` where `N` is the number of samples and `D` is the number of dimensions, but got tensor of shape )rndim ValueErrorshapefloat)rrr q/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/torchmetrics/functional/regression/cosine_similarity.py_cosine_similarity_updates rsum reductioncCsZ||jdd}|jdd}|jdd}|||}tjtjddddd}|||S)a&Compute Cosine Similarity. Args: preds: Predicted tensor target: Ground truth tensor reduction: The method of reducing along the batch dimension using sum, mean or taking the individual scores Example: >>> target = torch.tensor([[1, 2, 3, 4], [1, 2, 3, 4]]) >>> preds = torch.tensor([[1, 2, 3, 4], [-1, -2, -3, -4]]) >>> preds, target = _cosine_similarity_update(preds, target) >>> _cosine_similarity_compute(preds, target, 'none') tensor([ 1.0000, -1.0000]) )dimcS|SNr xr r rEz,_cosine_similarity_compute..cSrrr rr r rrFr)rmeannoneN)rnormtorchr)rrr dot_product preds_norm target_norm similarityreduction_mappingr r r_cosine_similarity_compute-s    r#cCst||\}}t|||S)aCompute the `Cosine Similarity`_. .. math:: cos_{sim}(x,y) = \frac{x \cdot y}{||x|| \cdot ||y||} = \frac{\sum_{i=1}^n x_i y_i}{\sqrt{\sum_{i=1}^n x_i^2}\sqrt{\sum_{i=1}^n y_i^2}} where :math:`y` is a tensor of target values, and :math:`x` is a tensor of predictions. Args: preds: Predicted tensor with shape ``(N,d)`` target: Ground truth tensor with shape ``(N,d)`` reduction: The method of reducing along the batch dimension using sum, mean or taking the individual scores Example: >>> from torchmetrics.functional.regression import cosine_similarity >>> target = torch.tensor([[1, 2, 3, 4], ... [1, 2, 3, 4]]) >>> preds = torch.tensor([[1, 2, 3, 4], ... [-1, -2, -3, -4]]) >>> cosine_similarity(preds, target, 'none') tensor([ 1.0000, -1.0000]) )rr#)rrrr r rcosine_similarityKs r$)r) typingrrrtorchmetrics.utilities.checksrtuplerstrr#r$r r r rs     $