Does anyone know how to stop an optimisation that I run with lbfgs_grad_free? I have set "maxiter"=1000, "gtol"= 1e-3, but the optimiser goes way beyond that and doesn't stop.
Using maxfun as an option leads to:
No, that gives "Unknown solver options: maxfun"
To Reproduce
Steps to reproduce the behavior. This can be in the form of a code snippet or some config files or a notebook to produce a minimum viable example of the bug.
Expected behavior
A clear and concise description of what you expect(ed) to happen.
Screenshots
If applicable, add screenshots to help explain your problem.
Environment (please complete the following information)
c3-toolset Version [e.g. 1.4]
Additional context
From @nwittler:
lbfgs has a maxfun option. The code for lbfgs_grad_free however is just minimize(fun=fun, x0=x_init, options=options). According to the docs, this could result in several methods being used. Try explicitly method="L-BFGS-B" with the maxfun option.
Describe the bug
From @alex-simm:
Using
maxfunas an option leads to:To Reproduce
Steps to reproduce the behavior. This can be in the form of a code snippet or some config files or a notebook to produce a minimum viable example of the bug.
Expected behavior
A clear and concise description of what you expect(ed) to happen.
Screenshots
If applicable, add screenshots to help explain your problem.
Environment (please complete the following information)
c3-toolsetVersion [e.g. 1.4]Additional context
From @nwittler: