Hi, I'm running
mert-moses-new.pl ~/en-fr_chart/tuning/token_lowercase.en ~/en-fr_chart/tuning/token_lowercase.fr /export/home/moses/libexec/moses-chart/bin/moses_chart ~/en-fr_chart/training/pm/model/moses.ini --mertdir=/export/home/moses/libexec/moses-chart/bin --working-dir=~/en-fr_chart/tuning --decoder-flags=-v 0 --no-filter-phrase-table for tuning. It performs exactly one iteration and writes a run1.moses.ini in the tuning directory. That moses.ini is almost identical to the one that I got after training (see below). Am I missing the obvious? any hints are welcome Christof [mo...@canaan tuning]$ diff -wc run1.moses.ini ../training/pm/model/moses.ini *** run1.moses.ini 2010-01-27 16:11:08.000000000 -0800 --- ../training/pm/model/moses.ini 2010-01-27 15:58:29.000000000 -0800 *************** *** 1,8 **** ! # MERT optimized configuration ! # decoder /export/home/moses/libexec/moses-chart/bin/moses_chart ! # BLEU --not-estimated-- on dev /export/home/moses/data/training/en-fr_chart/tuning/token_lowercase.en ! # We were before running iteration 1 ! # finished Wed Jan 27 16:11:08 PST 2010 ### MOSES CONFIG FILE ### ######################### --- 1,4 ---- ! ######################### ### MOSES CONFIG FILE ### ######################### *************** *** 35,41 **** # language model weights [weight-l] ! 1 # translation model weights --- 31,37 ---- # language model weights [weight-l] ! 0.5000 # translation model weights *************** *** 51,57 **** # word penalty [weight-w] ! 0 [cube-pruning-pop-limit] 1000 --- 47,53 ---- # word penalty [weight-w] ! -1 [cube-pruning-pop-limit] 1000 *************** *** 71,79 **** [max-chart-span] 20 1000 - - [v] - 0 - - [weight-d] - 1 --- 67,69 ---- _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
