Classification Model<T>.Predict Probabilities Core Method
Definition
Namespace: Numerics.NET.DataAnalysis.Models
Assembly: Numerics.NET (in Numerics.NET.dll) Version: 9.0.3
Assembly: Numerics.NET (in Numerics.NET.dll) Version: 9.0.3
Overload List
Predict | Predicts the probabilities of each class based on the specified values of the features. |
Predict | Predicts the probabilities of each class based on the specified values of the features. |
PredictProbabilitiesCore(Matrix<T>, Matrix<Double>, Boolean)
Predicts the probabilities of each class based on
the specified values of the features.
protected virtual Matrix<double> PredictProbabilitiesCore(
Matrix<T> features,
Matrix<double>? probabilities,
bool modelFeaturesOnly
)
Parameters
- features Matrix<T>
- A matrix containing the values for the independent variables.
- probabilities Matrix<Double>
- A matrix that is to hold the result.
- modelFeaturesOnly Boolean
- Specifies whether features includes only features that appear in the fitted model.
Return Value
Matrix<Double>A matrix whose rows contain the probabilities that the observation in the corresponding row of features belongs to each class.
Exceptions
Argument | features is null. |
Dimension | The number of columns of features does not match the required number of values. |
PredictProbabilitiesCore(Vector<T>, Vector<Double>, Boolean)
Predicts the probabilities of each class based on
the specified values of the features.
protected abstract Vector<double> PredictProbabilitiesCore(
Vector<T> features,
Vector<double>? probabilities,
bool modelFeaturesOnly
)
Parameters
- features Vector<T>
- A vector containing the values for the independent variables.
- probabilities Vector<Double>
- The vector that is to hold the result. May be null.
- modelFeaturesOnly Boolean
- Specifies whether features includes only features that appear in the fitted model.
Return Value
Vector<Double>The probabilities that the observation in features belongs to each class.
Exceptions
Argument | features is null. |
Dimension | The length of features does not match the required number of values. |