P-Matrices have obscenely large values when using a single rate category on the attached dataset

[GoogleCodeExporter]

Run the command:
./src/phyml-beagle -i ./datasets/fourtaxon_mod.dat -d nt -q -c 1 -v 0 -t e -m 
JC69 -f '0.25,0.25,0.25,0.25' -o none -b 0 --r_seed 1999
Observe that the tree is fixed, with no parameter optimizations (i.e. -o none) 
and we use one rate category (i.e. -c 1).
The P-matrix have very large values in their cells (possibly hitting machine 
precision limits?). I cant tell (but look at gdb_out.txt)

You can recreate the result with:
gdb -x ./phyml-beagle.gdbinit -batch --args ./src/phyml-beagle -i 
./datasets/fourtaxon_mod.dat -d nt -q -c 1 -v 0 -t e -m JC69 -f 
'0.25,0.25,0.25,0.25' -o none -b 0 --r_seed 1999 > ~/gdb_out.txt

Original issue reported on code.google.com by imran.fa...@gmail.com on 28 Jun 2013 at 10:24

Attachments:

• fourtaxon_mod.dat
• gdb_out.txt
• phyml-beagle.gdbinit

[GoogleCodeExporter]

Fixed in r1165  

Original comment by stephane.guindon on 18 Sep 2013 at 10:06

• Removed labels: PhyML-BEAGLE

[GoogleCodeExporter]

The change for this is done now.

Original comment by stephane.guindon on 26 Sep 2013 at 11:34

• Changed state: Fixed