16 Jan
2020
16 Jan
'20
11:53 a.m.
Am 16.01.20 um 10:45 schrieb Davide Cittaro:
Hello Tiago, thanks for the answer
On 16 Jan 2020, at 10:40, Tiago de Paula Peixoto <tiago@skewed.de> wrote:
The difference between the epsilon in mcmc_equilibrate() and in hiearchy_minimize() is explained in the documentation.
Indeed, but I wanted to know how the epsilon in the hiearchy_minimize affects speed. I guess I'll run some tests. Best,
No, it will not. From the documentation: epsilon: ``float`` (optional, default: ``1e-8``) Only replace levels if the description length difference is above this threshold. This is about numerical precision, nothing else. Most of the time you should leave this parameter alone. -- Tiago de Paula Peixoto <tiago@skewed.de>