search

khabbazian / l1ou

Detection of evolutionary shifts in Ornstein-Uhlenbeck models
GNU General Public License v2.0
19 stars 7 forks source link

issue with candidate edges #49

Closed jakeberv closed 2 years ago

jakeberv commented 2 years ago

I've been running a few experiments with l1ou and it seems like analysis of multiple traits takes much longer when candidate edges are specified (ie, a total set of edges that is less than all edges in the tree). Is this expected?

I'm currently running a particularly extreme test with 8 traits, 200 tips -- the model is using about 70gb of memory on the first LASSO (and > 20 hours of computing so far). Running the same model without candidate edges completes in less time.

Thanks, Jake Berv

jakeberv commented 2 years ago

I think something was going wrong in my R session -- I started over and this completed quickly. Not sure what happened.