function model = deltablssvm(model,a1,a2)% Bias term correction for the LS-SVM classifier% % >> model = deltablssvm(model, b_new)% % This function is only useful in the object oriented function% interface. Set explicitly the bias term b_new of the LS-SVM model.% % Full syntax% % >> model = deltablssvm(model, b_new)% % Outputs    %   model : Object oriented representation of the LS-SVM model with initial hyperparameters% Inputs    %   model : Object oriented representation of the LS-SVM model%   b_new : m x 1 vector with new bias term(s) for the model% % See also:%   roc, trainlssvm, simlssvm, changelssvm% Copyright (c) 2002,  KULeuven-ESAT-SCD, License & help @ http://www.esat.kuleuven.ac.be/sista/lssvmlabif iscell(model),  model = initlssvm(model{:});endif iscell(a1),  model.alpha = a1{1};  model.b = a1{2};  model.status = 'trained';  deltab = a2;else  deltab = a1;endif ~(model.type(1)=='c' & model.y_dim==1),  error('only for binary classification tasks');end% without retrainingmodel.b = deltab;function model = deltablssvm(model,a1,a2)% Set the bias of the binary classification model%% When training the LS-SVM in a standard way, no prior information% is incorporated. There exists however techniques who can% calculate a bias term according to another criterium. This% function allows to set the corrected bias in the final LS-SVM model. %%   model = deltablssvm(model,newbias)%   model = deltablssvm({X,Y,'classification',gam,sig2},{alpha,b},newbias)%% see also:%   roc, changelssvm% copyrightif iscell(model),  model = initlssvm(model{:});endif iscell(a1),  model.alpha = a1{1};  model.b = a1{2};  model.status = 'trained';  deltab = a2;else  deltab = a1;endif ~(model.type(1)=='c' & model.y_dim==1),  error('only for binary classification tasks');end% without retrainingmodel.b = deltab;