MetricType

Enums · AiDotNet.Enums

Defines the types of metrics used to evaluate machine learning models.

Fields

Field Summary
MetricType AIC A measure of the relative quality of statistical models for a given set of data.
MetricType AICAlt Alternative Akaike Information Criterion - A variant of AIC with a different penalty term.
MetricType AUC Alias of AUCROC.
MetricType AUCPR Area Under the Precision-Recall Curve - Measures classification accuracy focusing on positive cases.
MetricType AUCROC Area Under the Receiver Operating Characteristic Curve - Measures classification accuracy across thresholds.
MetricType AbsoluteRelativeError Depth error metrics for depth estimation.
MetricType Accuracy The proportion of correct predictions among all predictions made.
MetricType AdjustedR2 A modified version of R² that accounts for the number of predictors in the model.
MetricType AdjustedRandIndex Measures the similarity between two clusterings adjusted for chance.
MetricType AverageEpisodeReward The average total reward per episode (reinforcement learning).
MetricType AveragePrecision Measures the average precision across different recall levels.
MetricType BIC A criterion for model selection that penalizes model complexity more strongly than AIC.
MetricType BhattacharyyaDistance Measures the similarity between two probability distributions.
MetricType BrierScore A proper score function that measures the accuracy of probabilistic predictions.
MetricType CRPS Continuous Ranked Probability Score for probabilistic forecasts.
MetricType CalibrationError Measures how well the predicted probabilities match the actual outcomes.
MetricType CalinskiHarabaszIndex A measure of how similar an object is to its own cluster compared to other clusters.
MetricType ChamferDistance Chamfer Distance for point cloud similarity measurement.
MetricType CohenKappa Measures the agreement between two sets of labels.
MetricType ConditionNumber Measures the sensitivity of a system to small changes in input.
MetricType CosineSimilarity A measure of similarity between two non-zero vectors.
MetricType CrossEntropyLoss Measures the average log-loss across all classes in classification problems.
MetricType DaviesBouldinIndex A metric for evaluating clustering algorithms.
MetricType DurbinWatsonStatistic Durbin-Watson Statistic - Detects autocorrelation in prediction errors.
MetricType EarthMoversDistance Earth Mover’s Distance (Wasserstein) for point cloud comparison.
MetricType EffectiveNumberOfParameters The effective number of parameters in a model, useful for model complexity assessment.
MetricType EuclideanDistance The straight-line distance between two points in Euclidean space.
MetricType ExplainedVarianceScore Measures the proportion of variance in the dependent variable explained by the model.
MetricType F1Score The harmonic mean of precision and recall, providing a balance between the two metrics.
MetricType F2Score Measures the harmonic mean of precision and recall, with more weight on recall.
MetricType FBetaScore Measures the weighted harmonic mean of precision and recall.
MetricType FirstQuartile The first quartile (25th percentile) of the dataset.
MetricType GMean Measures the geometric mean of precision and recall.
MetricType HammingDistance The number of positions at which the corresponding symbols are different.
MetricType InterquartileRange The range of the middle 50% of the data.
MetricType IoU3D Intersection over Union for 3D volumetric segmentation.
MetricType JaccardSimilarity A statistic used for comparing the similarity and diversity of sample sets.
MetricType KLDivergence Measures the difference between two probability distributions.
MetricType KendallTau Measures the ordinal association between predicted and actual values.
MetricType Kurtosis Measures the “tailedness” of the probability distribution of a real-valued random variable.
MetricType LPIPS Learned Perceptual Image Patch Similarity for deep perceptual quality.
MetricType LevenshteinDistance Measures the similarity between two sequences.
MetricType Likelihood A measure of how probable the observed data is under the model.
MetricType LogLikelihood The natural logarithm of the likelihood, used for numerical stability and easier interpretation.
MetricType LogPointwisePredictiveDensity A measure of the model’s predictive accuracy on a point-by-point basis.
MetricType MAD The median absolute deviation (MAD) of the dataset.
MetricType MAE Mean Absolute Error - The average absolute difference between predicted and actual values.
MetricType MAPE Mean Absolute Percentage Error - The average percentage difference between predicted and actual values.
MetricType MSE Mean Squared Error - The average of squared differences between predicted and actual values.
MetricType MahalanobisDistance A multi-dimensional generalization of measuring how many standard deviations away a point is from the mean of a distribution.
MetricType ManhattanDistance The sum of the absolute differences of the coordinates.
MetricType MarginalLikelihood The probability of the observed data under all possible parameter values.
MetricType Max The largest value in the dataset.
MetricType MaxError Measures the maximum error between predicted and actual values.
MetricType Mean The arithmetic average of a set of values.
MetricType MeanAbsoluteError The average of the absolute differences between predicted and actual values.
MetricType MeanAbsolutePercentageError The average of the absolute percentage differences between predicted and actual values.
MetricType MeanAveragePrecision The mean of the average precision scores for each query.
MetricType MeanBiasError Mean Bias Error - The average of prediction errors (predicted - actual).
MetricType MeanIoU Mean Intersection over Union across all classes.
MetricType MeanPredictionError The average difference between predicted values and actual values.
MetricType MeanReciprocalRank The average of the reciprocal ranks of the first relevant items.
MetricType MeanSquaredError The average of the squared differences between predicted and actual values.
MetricType MeanSquaredLogError Mean Squared Logarithmic Error - Penalizes underestimates more than overestimates.
MetricType Median The middle value in a sorted list of numbers.
MetricType MedianAbsoluteError Median Absolute Error - The middle value of all absolute differences between predicted and actual values.
MetricType MedianPredictionError The middle value of all differences between predicted values and actual values.
MetricType Min The smallest value in the dataset.
MetricType Mode The most frequently occurring value in a dataset.
MetricType MutualInformation Measures the mutual dependence between two variables.
MetricType N The number of values in the dataset.
MetricType NormalConsistency Normal consistency for surface reconstruction quality.
MetricType NormalizedDiscountedCumulativeGain A measure of ranking quality used in information retrieval.
MetricType NormalizedMutualInformation A normalization of the Mutual Information to scale the results between 0 and 1.
MetricType ObservedTestStatistic The value of the test statistic computed from the observed data.
MetricType PSNR Peak Signal-to-Noise Ratio for rendered image quality.
MetricType PartIoU Part-averaged IoU for part segmentation tasks.
MetricType PearsonCorrelation Measures the linear correlation between predicted and actual values.
MetricType Perplexity Measures the log-likelihood of held-out data under a topic model.
MetricType PopulationStandardError Population Standard Error - The standard deviation of prediction errors without adjustment for model complexity.
MetricType Precision The proportion of true positive predictions among all positive predictions.
MetricType PredictionIntervalCoverage The percentage of actual values that fall within the model’s prediction intervals.
MetricType R2 Coefficient of determination, measuring how well the model explains the variance in the data.
MetricType RMSE Root Mean Squared Error - The square root of the Mean Squared Error.
MetricType ROCAUCScore Measures the trade-off between true positive rate and false positive rate.
MetricType RSquared Alias of R2.
MetricType Range The difference between the largest and smallest values in a dataset.
MetricType Recall The proportion of true positive predictions among all actual positives.
MetricType ReferenceModelMarginalLikelihood The marginal likelihood of a reference or null model.
MetricType RootMeanSquaredError The square root of the mean squared error.
MetricType SMAPE Symmetric Mean Absolute Percentage Error - A variant of MAPE that handles zero or near-zero values better.
MetricType SSIM Structural Similarity Index for perceptual image quality.
MetricType SampleStandardError Sample Standard Error - An estimate of the standard deviation of prediction errors, adjusted for model complexity.
MetricType SilhouetteScore Measures the quality of clustering algorithms.
MetricType Skewness Measures the asymmetry of the probability distribution of a real-valued random variable.
MetricType SpearmanCorrelation Measures the monotonic relationship between predicted and actual values.
MetricType StandardDeviation The square root of the variance, measuring the average deviation from the mean.
MetricType TheilUStatistic Theil’s U Statistic - A measure of forecast accuracy relative to a naive forecasting method.
MetricType ThirdQuartile The third quartile (75th percentile) of the dataset.
MetricType Variance A measure of variability in the data, calculated as the average squared deviation from the mean.
MetricType VariationOfInformation A measure of the amount of information lost when compressing two random variables.
int value__