Crate tangram_metrics[][src]

Expand description

This crate implements a number of metrics such as MeanSquaredError and Accuracy.

Structs

The accuracy is the proportion of examples where predicted == label.

This Metric computes the area under the receiver operating characteristic curve.

BinaryClassificationMetrics computes common metrics used to evaluate binary classifiers at a number of classification thresholds.

BinaryClassificationMetrics contains common metrics used to evaluate binary classifiers.

The output from BinaryClassificationMetrics.

BinaryCrossEntropy is the loss function used in binary classification. Learn more.

The input to BinaryCrossEntropy.

ClassMetrics are class specific metrics used to evaluate the model’s performance on each individual class.

CrossEntropy is the loss function used in multiclass classification. Learn more.

The input to CrossEntropy.

The output from CrossEntropy.

The Mean metric is computed using Welford’s algorithm to ensure numeric stability.

The mean squared error is the sum of squared differences between the predicted value and the label.

MulticlassClassificationMetrics computes common metrics used to evaluate multiclass classifiers.

RegressionMetrics computes metrics used to evaluate regressors.

The input to RegressionMetrics.

The output from RegressionMetrics.

Functions

This function computes the variance given the m2 and n.

This function combines two separate means and variances into a single mean and variance which is useful in parallel algorithms.