search

lutteropp / NetRAX

Phylogenetic Network Inference without ILS
GNU General Public License v3.0
17 stars 1 forks source link

When and in which order optimize reticulation probs, branch lengths, and model parameters? #37

Open lutteropp opened 3 years ago

lutteropp commented 3 years ago

Turns out the order of {reticulationOpt, brlenOpt, modelOpt}, and when and how often to call it during the search is affecting which moves are getting accepted.

lutteropp commented 3 years ago

Reopened because this is now the most pressing issue: If we do exhaustive {reticulationOpt, brlenOpt, modelOpt}, NetRAX becomes unbearably slow. If we don't do enough, it won't accept any moves...

stamatak commented 3 years ago

slow is not such an issue, we are really interested in the concept here

On 12.01.21 11:31, Sarah Lutteropp wrote:

Reopened because this is now the most pressing issue: If we do exhaustive {reticulationOpt, brlenOpt, modelOpt}, NetRAX becomes unbearably slow. If we don't do enough, it won't accept any moves...

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/lutteropp/NetRAX/issues/37#issuecomment-758528389, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAGXB6Q3SH2S47PQR2N7KA3SZQJH7ANCNFSM4V5U4RPQ.

-- Alexandros (Alexis) Stamatakis

Research Group Leader, Heidelberg Institute for Theoretical Studies Full Professor, Dept. of Informatics, Karlsruhe Institute of Technology

www.exelixis-lab.org