Skip to main content
. 2019 Feb 1;36(4):825–833. doi: 10.1093/molbev/msz020

Table 1.

Run Times and Estimated Trees of Variational Inference and MCMC Algorithms on Real Data.

Data Set Taxa Sites States MCMC (days) VI (days)
Data Set A 13 6,622 20 4.72 0.81
Data Set B 28 10,137 20 10.61 2.36
Data Set C 66 38,330 20 28.35 5.67

Note.—Both variational inference and MCMC algorithms were run in a parallel environment. The properties of the parallel version were evaluated on a personal computer (Intel Core i7-6700 CPU 3.40 GHz, 8 cores, 2 threads per core, 4 cores per socket, 16-Gb RAM), under Linux Mint 17.3 Rosa. In this comparative study, both variational inference and MCMC had 30,000 iterations (see text).