Home | History | Annotate | Download | only in ceres

Lines Matching refs:function

35 // To get a numerically differentiated cost function, define a subclass of
36 // CostFunction such that the Evaluate() function ignores the jacobian
38 // parameter if nececssary by repeatedly calling the Evaluate() function with
41 // concrete cost function, even though it could be implemented only in terms of
44 // The numerically differentiated version of a cost function for a cost function
55 // twice as many function evaluations than forward difference. Consider using
89 const CostFunctionNoJacobian *function,
142 if (!function->Evaluate(parameters, residuals, NULL)) {
156 // Compute the function on the other side of x(j).
159 if (!function->Evaluate(parameters, residuals, NULL)) {
191 const CostFunctionNoJacobian *function,
206 NumericDiffCostFunction(CostFunctionNoJacobian* function,
208 : function_(function), ownership_(ownership) {}
219 // Get the function value (residuals) at the the point to evaluate.