Metrics¶
Aggregation Metrics¶
Computes Area Under the Curve (AUC) using the trapezoidal rule. |
|
Concatenate all input tensors along dimension dim. |
|
Calculate the maximum value of all elements in all the input tensors. |
|
Calculate the weighted mean value of all elements in all the input tensors. |
|
Calculate the minimum value of all elements in all the input tensors. |
|
Calculate the weighted sum value of all elements in all the input tensors. |
|
Calculate the throughput value which is the number of elements processed per second. |
Classification Metrics¶
Compute binary accuracy score, which is the frequency of input matching target. |
|
Compute AUPRC, also called Average Precision, which is the area under the Precision-Recall Curve, for binary classification. |
|
Compute AUROC, which is the area under the ROC Curve, for binary classification. |
|
Compute AUROC, which is the area under the ROC Curve, for binary classification. |
|
Compute precision recall curve with given thresholds. |
|
Compute binary confusion matrix, a 2 by 2 tensor with counts ( (true positive, false negative) , (false positive, true negative) ) |
|
Compute binary f1 score, which is defined as the harmonic mean of precision and recall. |
|
Compute the normalized binary cross entropy between predicted input and ground-truth binary target. |
|
Compute the precision score for binary classification tasks, which is calculated as the ratio of the true positives and the sum of true positives and false positives. |
|
Returns precision-recall pairs and their corresponding thresholds for binary classification tasks. |
|
Compute the recall score for binary classification tasks, which is calculated as the ratio of the true positives and the sum of true positives and false negatives. |
|
Returns the highest possible recall value give the minimum precision for binary classification tasks. |
|
Compute accuracy score, which is the frequency of input matching target. |
|
Compute AUPRC, also called Average Precision, which is the area under the Precision-Recall Curve, for multiclass classification. |
|
Compute AUROC, which is the area under the ROC Curve, for multiclass classification in a one vs rest fashion. |
|
Compute AUROC, which is the area under the ROC Curve, for multiclass classification. |
|
Compute precision recall curve with given thresholds. |
|
Compute multi-class confusion matrix, a matrix of dimension num_classes x num_classes where each element at position (i,j) is the number of examples with true class i that were predicted to be class j. |
|
Compute f1 score, which is defined as the harmonic mean of precision and recall. |
|
Compute the precision score, the ratio of the true positives and the sum of true positives and false positives. |
|
Returns precision-recall pairs and their corresponding thresholds for multi-class classification tasks. |
|
Compute the recall score, the ratio of the true positives and the sum of true positives and false negatives. |
|
Compute multilabel accuracy score, which is the frequency of input matching target. |
|
Compute AUPRC, also called Average Precision, which is the area under the Precision-Recall Curve, for multilabel classification. |
|
Returns precision-recall pairs and their corresponding thresholds for multi-label classification tasks. |
|
Returns the highest possible recall value given the minimum precision for each label and their corresponding thresholds for multi-label classification tasks. |
|
Compute multilabel accuracy score, which is the frequency of the top k label predicted matching target. |
Ranking Metrics¶
Compute the click through rate given click events. |
|
Compute the hit rate of the correct class among the top predicted classes. |
|
Compute the reciprocal rank of the correct class among the top predicted classes. |
|
Compute weighted calibration metric. |
Regression Metrics¶
Compute Mean Squared Error, which is the mean of squared error of input and target. |
|
Compute R-squared score, which is the proportion of variance in the dependent variable that can be explained by the independent variable. |
Text Metrics¶
Compute BLEU score (https://en.wikipedia.org/wiki/BLEU) given translations and references. |
|
Perplexity measures how well a model predicts sample data. |
|
Compute the word error rate of the predicted word sequence(s) with the reference word sequence(s). |
|
Word Information Lost (WIL) is a metric of the performance of an automatic speech recognition system. |
|
Compute the word information preserved of the predicted word sequence(s) with the reference word sequence(s). |
Windowed Metrics¶
The windowed version of BinaryAUROC. |
|
The windowed version of BinaryNormalizedEntropy that provides both windowed and liftime values. |
|
The windowed version of ClickThroughRate that provides both windowed and lifetime values. |
|
The windowed version of Mean Squared Error that provides both windowed and liftime values. |
|
Compute weighted calibration metric. |