[SciPy-User] fast small matrix multiplication with cython?

josef.pktd@gmai... josef.pktd@gmai...
Wed Dec 8 22:28:13 CST 2010


>
> It looks like I don't save too much time with just Python/scipy
> optimizations.  Apparently ~75% of the time is spent in l-bfgs-b,
> judging by its user time output and the profiler's CPU time output(?).
>  Non-cython versions:
>
> Brief and rough profiling on my laptop for ARMA(2,2) with 1000
> observations.  Optimization uses fmin_l_bfgs_b with m = 12 and iprint
> = 0.

Completely different idea: How costly are the numerical derivatives in l-bfgs-b?
With l-bfgs-b, you should be able to replace the derivatives with the
complex step derivatives that calculate the loglike function value and
the derivatives in one iteration.

Josef


More information about the SciPy-User mailing list