July 2012 Commit Log

Number of Commits:
18
Number of Active Developers:
8
rambaut 2012-07-31 03:25 Rev.: 5078

Fixed issue with precision bounds for dates of leaf nodes

1 lines of code changed in 1 file:

jheled@gmail.com 2012-07-23 14:13 Rev.: 5077

Improve report sent by BEAST when initial posterior is infinite

44 lines of code changed in 4 files:

rambaut 2012-07-21 03:56 Rev.: 5076

Precision field in date now seems to work correctly: if you give a precision="1.0" to a date then it assumes that the date was sampled from the given t to t+1 in whatever units are being used (forwards and backwards dates are accommodated). If a treemodel leafHeight is created then this range is automatically added as a bound allowing a randomwalk operator to be used.

2 lines of code changed in 1 file:

fbielejec 2012-07-19 07:46 Rev.: 5075

exception handling

226 lines of code changed in 8 files:

rambaut 2012-07-19 04:52 Rev.: 5074

Added an option to define a precision level for a date which automatically creates appropriate bounds for a leaf-height parameter. Then you can sample from those bounds using a uniform or randomwalk operator. The attribute: precision="1.0" means the date has been specified to the precision of 1 unit (missing or 0.0 means exact precision).

12 lines of code changed in 1 file:

rambaut 2012-07-19 04:41 Rev.: 5073

Overhauled the date system so that the 'height' timescale is now universal to all taxa. Previously there was an issue if you create independent trees on different taxon sets, they would have their own 'zero' point depending on their most recently sampled taxon. Now, when you call 'setDate' for a taxon (as done by the TaxonParser) it keeps track of the globally most recent sample (in a static variable) and can give the global height from a given date object using a static function. Taxon objects also have a 'getHeight' methods that uses this system. The two main sources of trees (CoalescentSimulator and NewickParser) now use these heights (simplifying their own code). Runs benchmarks correctly but keep an eye out for issues.

152 lines of code changed in 8 files:

fbielejec 2012-07-18 12:25 Rev.: 5072

working on epoch tab

200 lines of code changed in 5 files:

graham853@gmail.com 2012-07-17 02:28 Rev.: 5071

Code for n diploids, one hybridization now working. Experimental code (PopsIO) added for multispecies coalescent with population size parameters integrated out.

2393 lines of code changed in 35 files:

rambaut 2012-07-17 01:52 Rev.: 5070

Cleaning up a few issues with the Mac L&F

9 lines of code changed in 3 files:

rambaut 2012-07-17 00:28 Rev.: 5069

Put the target JDK version as a property at the top of the ant file.

3 lines of code changed in 1 file:

akaruiws 2012-07-16 17:56 Rev.: 5068

Provide more realistic default value for model parameter just like providing more realistic starting tree to the analysis to prevent numerical instability or whatever silly stuff that does not allow the analysis to start running.

1 lines of code changed in 2 files:

alexei.drummond 2012-07-16 15:55 Rev.: 5067

fixed microsatellite initial value errors in geomDist and propLinear parameters (if in doubt, shouldn't start on boundary values as a general rule).

2 lines of code changed in 1 file:

fbielejec 2012-07-14 03:24 Rev.: 5066

working on epoch model tab

11 lines of code changed in 3 files:

fbielejec 2012-07-13 05:35

working on bss UI

273 lines of code changed in 4 files:

msuchard 2012-07-12 05:50 Rev.: 5063

Default behavior brings up GUI panel instead of just the open file panel

344 lines of code changed in 3 files:

msuchard 2012-07-10 15:52 Rev.: 5062

Fixed more -1s in thread-pools and add limited dialogue support for SSE (which sidesteps known bug in SSE phylogeography)

61 lines of code changed in 3 files:

gcybis@yahoo.com.br 2012-07-09 20:35 Rev.: 5061

Better mixing for hierarchical graph model

20 lines of code changed in 4 files:

rambaut 2012-07-07 04:08 Rev.: 5060

Spelling.

1 lines of code changed in 2 files:

June 2012 »

Generated by StatSVN 0.7.0