fann_get_cascade_output_change_fraction
The cascade output change fraction is a number between 0 and 1 determining how large a fraction of the fann_get_MSE value should change within fann_get_cascade_output_stagnation_epochs during training of the output connections, in order for the training not to stagnate. If the training stagnates, the training of the output connections will be ended and new candidates will be prepared.
It means that if the MSE does not change by a fraction of fann_get_cascade_output_change_fraction during a period of fann_get_cascade_output_stagnation_epochs, the training of the output connections is stopped because the training has stagnated.
If the cascade output change fraction is low, the output connections will be trained more and if the fraction is high, they will be trained less.
The default cascade output change fraction is 0.01, which is equalent to a 1% change in MSE.
Parameters
- ann
-
Neural network resource.
Return Values
The cascade output change fraction, or false on error.