The HiddenResilientGradientLearningT generic type exposes the following members.
Gets or sets the number of performed iterations.
Gets the decrease parameter, also referred as eta minus. Default is 0.5.
Gets or sets the potential function to be used if this learning algorithm needs to create a new HiddenConditionalRandomFieldT.(Inherited from BaseHiddenConditionalRandomFieldLearningT.)
Gets or sets whether the algorithm has converged.
Gets the increase parameter, also referred as eta plus. Default is 1.2.
|Iterations|| Obsolete. |
Please use MaxIterations instead.
Gets or sets the maximum number of iterations performed by the learning algorithm.
Gets or sets the model being trained.(Inherited from BaseHiddenConditionalRandomFieldLearningT.)
Gets or sets the parallelization options for this algorithm.
Gets or sets the amount of the parameter weights which should be included in the objective function. Default is 0 (do not include regularization).
Gets or sets a value indicating whether this HiddenGradientDescentLearningT should use stochastic gradient updates. Default is true.
Gets or sets a cancellation token that can be used to stop the learning algorithm while it is running.(Inherited from BaseHiddenConditionalRandomFieldLearningT.)
Gets or sets the maximum change in the average log-likelihood after an iteration of the algorithm used to detect convergence.
Gets or sets the minimum possible update step, also referred as delta max. Default is 1e-6.
Gets or sets the maximum possible update step, also referred as delta min. Default is 50.