Прочитај на енглеском

Делите путем


TorchSharpCatalog.EvaluateObjectDetection Method

Definition

Evaluates scored object detection data.

C#
public static Microsoft.ML.TorchSharp.AutoFormerV2.ObjectDetectionMetrics EvaluateObjectDetection(this Microsoft.ML.MulticlassClassificationCatalog catalog, Microsoft.ML.IDataView data, Microsoft.ML.DataViewSchema.Column labelCol, Microsoft.ML.DataViewSchema.Column actualBoundingBoxColumn, Microsoft.ML.DataViewSchema.Column predictedLabelCol, Microsoft.ML.DataViewSchema.Column predictedBoundingBoxColumn, Microsoft.ML.DataViewSchema.Column scoreCol);

Parameters

catalog
MulticlassClassificationCatalog

The transform's catalog.

data
IDataView

IDataView with the data

labelCol
DataViewSchema.Column

Column that has the actual labels.

actualBoundingBoxColumn
DataViewSchema.Column

Column that has the actual bounding boxes.

predictedLabelCol
DataViewSchema.Column

Column that has the predicted labels.

predictedBoundingBoxColumn
DataViewSchema.Column

Column that has the predicted bounding boxes.

scoreCol
DataViewSchema.Column

Column that has the predicted score (confidence level).

Returns

Applies to

Производ Верзије
ML.NET 3.0.0, 4.0.0, Preview