Jacobian elements order about cppad HOT 4 CLOSED

Which routine are you calling to get the Jacobian (sparse Jacobians have a special storage) ?

from cppad.

I used some dynamic lib code generation with CppADCodeGen, so I call it this way :

    LinuxDynamicLib<Scalar> fkine_Linear_Lib(FKINE_LINEAR_EXT);
    std::unique_ptr<GenericModel<double>> fkine_linear_model = fkine_Linear_Lib.model("fkine_linear");
    jac_positions = fkine_linear_model->Jacobian(q);

from cppad.

Hello,

The method model->Jacobian(q); returns the entire dense Jacobian (m x n) in a vector.
It provides the elements in the same order as calling CppAD JacobianFor or JacobianRev:

jac[ i * n + j ] = \D{ F_i }{ x_j } ( x )
i = 0 , ... , m - 1
j = 0 , ... , n - 1

The documentation is missing for those methods. I'll add this information.

from cppad.

Great ! thank you very much for your help :)

from cppad.