17 lines
357 B
Mathematica
17 lines
357 B
Mathematica
|
function f = ipopt_callback_f(x,model_in)
|
||
|
|
|
||
|
|
global latest_x
|
||
|
|
global latest_df
|
||
|
|
|
||
|
|
persistent model
|
||
|
|
if nargin>1
|
||
|
|
model = model_in;
|
||
|
|
return
|
||
|
|
end
|
||
|
|
|
||
|
|
% We compute both function value and gradient. If pennon calls later to get
|
||
|
|
% gradient, we have saved in a global variable (yea, I know, globals suck)
|
||
|
|
x = x(:);
|
||
|
|
[f,latest_df] = fmincon_fun(x,model);
|
||
|
|
latest_x = x;
|