15.4 Test the OutputManager class

(Win version)

< 15.3 | 15.4 | 16.0 >

Run the program using the following strom.conf file:

datafile    = rbcl10.nex    
treefile    = rbcl10.tre
statefreq   = default:0.294701,0.189205,0.205994,0.310100
rmatrix     = default:0.06082,0.27887,0.06461,0.06244,0.48492,0.04834
ratevar     = default:1.0
pinvar      = default:0.0
ncateg      = default:4
niter       = 1000
samplefreq  = 1
printfreq   = 100 
expectedLnL = -6897.492    

As before, the ncatag = default:4 is required to force the model to include rate heterogeneity, which in turn forces the gamma rate variance (ratevar) parameter to be updated during MCMC.

Here is the output you should expect:

Starting...
Current working directory: "/Users/plewis/Documents/websites/cpptutorial/steps/step-15/test"

*** Reading and storing the data in the file rbcl10.nex
storing read block: TAXA
storing read block: CHARACTERS

Number of taxa: 10
Number of partition subsets: 1
  Subset 1 (default)
    data type: nucleotide
    sites:     1314
    patterns:  424
    ambiguity: treated as missing data (faster)

*** Resources available to BeagleLib 3.1.2:
  resource 0: CPU

*** Creating the likelihood calculator

*** Model description
Partition information:

         data subset           1
   -----------------------------
          num. sites        1314
       num. patterns         424
         num. states           4
     rate categories           4

Parameter linkage:

         data subset           1
   -----------------------------
         state freqs           1
   exchangeabilities           1
               omega           -
       rate variance           1
              pinvar           -

Parameter values for each subset:

  relative rate:
             1: 1

  state freqs:
             1: (0.294701,0.189205,0.205994,0.3101)

  exchangeabilities:
             1: (0.06082,0.27887,0.06461,0.06244,0.48492,0.04834)

  omega:
             1: -

  rate variance:
             1: 1

  pinvar:
             1: -

Created BeagleLib instance 0 (4 states, 4 rates, 1 subset, not invar. sites model)

*** Reading and storing the first tree in the file rbcl10.tre

Warning:  
 A TAXA block should be read before the TREES block (but no TAXA block was found).  Taxa will be inferred from their usage in the TREES block.
at line 14, column (approximately) 1 (file position 346)
storing implied block: TAXA
storing read block: TREES

*** Calculating the likelihood of the tree
log likelihood = -6897.50119
      (expecting -6897.492)

   iteration      logLike     logPrior           TL
           0  -6897.50119     -1.00000      1.75014
         100  -6723.24797     -3.82676      1.75014
         200  -6723.36455     -3.77476      1.75014
         300  -6723.57824     -4.19035      1.75014
         400  -6725.66292     -3.39929      1.75014
         500  -6724.14492     -4.32369      1.75014
         600  -6724.08629     -3.60218      1.75014
         700  -6723.51995     -3.72464      1.75014
         800  -6723.80059     -3.65665      1.75014
         900  -6723.29172     -4.08572      1.75014
        1000  -6723.76580     -3.66414      1.75014

Chain summary:
                       Updater   Tuning Param.        Accept %     No. Updates
           Gamma Rate Variance           0.689            30.5            1000

Finished!

This output does not tell the whole story, however. To see that the gamma rate variance parameter (but no other parameter) is being updated, take a look at the params.txt file that was generated by the OutputManager object. Note that all columns remain at their starting values except the “iter” column, the log-likelihood (labeled “lnL”) column, the log-prior (labeled “lnPr”) column, and the last column labeled “ratevar-0” (the gamma rate variance parameter).

< 15.3 | 15.4 | 16.0 >