function varargout = entropy(varargin) %ENTROPY % % y = ENTROPY(x) % % Computes/declares entropy -sum(x.*log(x)) % % Implemented as evalutation based nonlinear operator. Hence, the concavity % of this function is exploited to perform convexity analysis and rigorous % modelling. % Author Johan Löfberg % $Id: gan.m,v 1.1 2006/08/17 07:57:24 joloef Exp $ switch class(varargin{1}) case 'double' % What is the numerical value of this argument (needed for displays etc) x = varargin{1}; a = varargin{2}; varargout{1} = sum(x(:).*a(:).^(1./x(:))); case 'sdpvar' % Overloaded operator for SDPVAR objects. X = varargin{1}; if min(size(X))>1 error('ENTROPY only defined for vector arguments'); else % y = []; % for i = 1:length(X) % y = [y;yalmip('addEvalVariable',mfilename,X(i),varargin{2})]; % end % varargout{1} = y; varargout{1} = yalmip('addEvalVariable',mfilename,varargin{:}); end case 'char' % YALMIP sends 'model' when it wants the epigraph or hypograph switch varargin{1} case 'graph' t = varargin{2}; X = varargin{3}; A = varargin{4}; % This is different from so called extended operators % Just do it! F = SetupEvaluationVariable(varargin{:}); % Now add your own code, such as domain constraints F = F + set(X > 0); % Let YALMIP know about convexity etc varargout{1} = F; varargout{2} = struct('convexity','convex','monotonicity','none','definiteness','none'); varargout{3} = X; case 'milp' varargout{1} = []; varargout{2} = []; varargout{3} = []; otherwise error('SDPVAR/LOG called with CHAR argument?'); end otherwise error('SDPVAR/LOG called with CHAR argument?'); end