Contents

convergenceThreshold

The convergence threshold.

Declaration

var convergenceThreshold: Double

Discussion

When the residual is within the convergence threshold of the objective, training stops. The threshold is also used by the fitted method to decide when progress is no longer being made, in which case the training process will stop before convergence and before the specified maximum number of iterations (known as early stopping).

Consider reducing this value for a more accurately trained model. But beware of overfitting if the it is set to a very low value. Defaults to 0.01.

See Also

Getting the properties