Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/moses-smt/mosesdecoder.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
path: root/mert
diff options
context:
space:
mode:
authorTetsuo Kiso <tetsuo-s@is.naist.jp>2012-04-08 17:45:04 +0400
committerTetsuo Kiso <tetsuo-s@is.naist.jp>2012-04-08 17:45:04 +0400
commit4c38bbb4343c7c9978d3f927abcacd467f788292 (patch)
tree65aa53c43c7174dbf840652f37b02d1e9a6a6a4f /mert
parentd034eeb703bd56776d0e83675ab4d105cda0e340 (diff)
Add things to TODO.
Diffstat (limited to 'mert')
-rw-r--r--mert/TODO18
1 files changed, 18 insertions, 0 deletions
diff --git a/mert/TODO b/mert/TODO
index 2559e78b5..028ca18e0 100644
--- a/mert/TODO
+++ b/mert/TODO
@@ -4,3 +4,21 @@
- this may make use of 'evaluator', soon to be added by Matous Machacek
- check that --pairwise-ranked is compatible with all optimization metrics
+
+- Replace the standard rand() currently used in MERT and PRO with better
+ random generators such as Boost's random generators (e.g., boost::mt19937).
+ - create a Random class to hide the details, i.e., how to generate
+ random numbers, which allows us to use custom random generators more
+ easily.
+
+ Pros:
+ - In MERT, you might want to use the random restarting technique to avoid
+ local optima.
+ - PRO uses a sampling technique to choose candidate translation pairs
+ from N-best lists, whic means the choice of random generators seems to
+ be important.
+
+ Cons:
+ - This change will require us to re-create the truth results for regression
+ testing related to MERT and PRO because the new random generator will
+ generate different numbers from the current generator does.