Joel Andersson
Joel Andersson
Cf. https://groups.google.com/forum/#!topic/casadi-users/V3yzkeYidkM
The support for convex optimization other than QP (LP, SDP, SDQP, SOCP, SOQP and QCQP) didn't survive the 3.0 refactoring. LP was later reintroduced as a special case of QP...
TO DOs for 3.6: - [ ] Re-enable Ipopt Callbacks - [ ] Populate change list: Add closed issues, new features [JA + JG] - [ ] FMU docs [JA]...
Extension of #2823. At least in a first iteration, the implementation will be based on Jacobian calculation only (not directional derivatives).
Derivative function (e.g. created using fcn.forward(1), fcn.reverse(1)) will have a set of options automatically derived from the generating class. For example, it makes little sense to propagate options such as...
Allow integrating general-purpose FMUs into CasADi expressions with derivative information exploited: - [x] Loading FMU DLLs - [x] Avoid reinitializing FMU when not necessary (i.e. when called inside an ODE...
A potentially useful atomic would be one that simply copies the results without change, but with zero derivatives, symbolic or numeric. It's trivial to implement symbolic evaluation, sparsity propagation and...
If I compile CasADi with SWIG version 4, i.e. the master branch of SWIG, trivial programs like these: ```python import casadi as ca y = lambda y: y x =...
The CasADi C wrappers appear to return references to temporary string objects. It shows up as a compiler warning for me: ``` /Users/jaeandersson/dev/casadi/casadi/core/casadi_c.cpp:222:10: warning: returning address of local temporary object...