Enum LossFunction
The loss function which determines how a prediction in the form of a distribution is converted into a point prediction.
Namespace: Microsoft.ML.Probabilistic.Learners
Assembly: Microsoft.ML.Probabilistic.Learners.dll
Syntax
public enum LossFunctionFields
| Name | Description |
|---|---|
| Absolute | The absolute loss function is equivalent to choosing the median of the predicted distribution as point estimate. Use this loss function to minimize mean absolute error. |
| Custom | The custom loss function allows to provide a user-defined loss function when converting a prediction in the form of a distribution into a point prediction. |
| Squared | The squared or quadratic loss function is equivalent to choosing the mean of the predicted distribution as point estimate. Use this loss function to minimize mean squared error. |
| ZeroOne | The zero-one loss function is equivalent to choosing the mode of the predicted distribution as point estimate. Use this loss function to minimize mean classification error. |