Note: This release is in preview. Features described here don’t become generally available until the latest general availability date that Salesforce announces for this release. Before then, and where features are noted as beta, pilot, or developer preview, we can’t guarantee general availability within any particular time frame or at all. Make your purchase decisions only on the basis of generally available products and features.

MLModelMetric

Represents a metric or statistic about the related model, such as accuracy, precision, or RSquared. Use a model’s metrics to learn about its performance and to compare it with other models. This object is available in API version 53.0 and later.

Supported Calls

describeSObjects(), getDeleted(), getUpdated(), query(), retrieve()

Special Access Rules

Available with Einstein Prediction Builder and Einstein Recommendation Builder.

Fields

Field	Details
BasicMetricValue	Type double Properties Filter, Nillable, Sort Description The value of a basic metric. A basic metric is a single number. For metrics that comprise a set of graph points, see ComplexMetricValue .
ComplexMetricValue	Type textarea Properties Nillable Description The X and Y values for a complex metric. A complex metric is a coordinate on a graph. For example, in classification models, you can use a line on a graph to create classification categories.
DataSetType	Type picklist Properties Filter, Group, Nillable, Restricted picklist, Sort Description The type of dataset. Possible values are: `Baseline` `HoldOut` `Live` `Model` `Training`
EndTime	Type dateTime Properties Filter, Nillable, Sort Description The date and time when the model training finished.
GraphType	Type picklist Properties Filter, Group, Nillable, Restricted picklist, Sort Description The type of graph. Possible values are: `ConfidencePlot` `ConfusionMatrixPerThreshold` `DiscountedCumulativeGainsGraph` `HitRateGraph` `KBasedRankingGraph` `LiftPlot` `MeanReciprocalRankGraph` `MultiClassConfusionMatrixPerThreshold` `MultiClassMisclassifications` `NormalizedDiscountedCumulativeGainsGraph` `PrecisionGraph` `RecallGraph` `RegressionErrorBands`
MetricType	Type picklist Properties Filter, Group, Nillable, Restricted picklist, Sort Description The type of metric. Possible values are: `Accuracy` `AveragePrecision` `BalancedAccuracy` `DiscountedCumulativeGainAtK` `ExpectedTopAbsoluteRank` `ExpectedTopPercentileRank` `F1Score` `FMeasure` `HitRateAtK` `LiftBucket` `MeanAbsoluteError` `MeanAbsoluteRank` `MeanAveragePrecisionAtK` `MeanPercentileRank` `MeanReciprocalRank` `MeanReciprocalRankAtK` `MeanTopReciprocalRank` `NormalizedDiscountedCumulativeGainsAtK` `Precision` `PrecisionAtK` `RSquared` `Recall` `RecallAtK` `RootMeanSquaredError` `auPR` `auROC`
ModelId	Type reference Properties Filter, Group, Sort Description The ID of the related MLModel. This field is a polymorphic relationship field. Relationship Name Model Relationship Type Lookup Refers To MLModel
Name	Type string Properties Autonumber, Defaulted on create, Filter, idLookup, Sort Description An automatically generated ID that uniquely identifies the metric.
RowCount	Type int Properties Filter, Group, Nillable, Sort Description The number of rows.
Span	Type picklist Properties Filter, Group, Nillable, Restricted picklist, Sort Description The time span for the metric. Possible values are: `Day` `Hour` `Month` `SinceLastAction` `Week`
StartTime	Type dateTime Properties Filter, Nillable, Sort Description The date and time when the model training started.

Object Reference for the Salesforce Platform

MLModelMetric

Supported Calls

Special Access Rules

Fields