[developers] questions about PET/ERG system

Stephan Oepen oe at ifi.uio.no
Wed Apr 8 18:31:52 CEST 2015

dear ping,

there is some documentation on the general unknown word handling
mechanisms in PET:


for the modern ERG, only the newer method is actively supported (and
tested in the past ten or so years), i.e. for most use cases of the
ERG one wants to enable REPP (string normalization and tokenization;
see ‘ReppTop’ on the wiki), token mapping, the modern, controlled way
of unknown word instantiation (‘default-les=all’), and lexical
filtering.  token mapping and lexical filtering, together, are often
called chart mapping (‘-cm’).  one easy-to-use way of using the ERG is
within the LOGON environment, see:


but it sounds as if you want a little more control than those
pre-defined configurations, maybe.  a good starting point for invoking
PET could be a command line along the following lines:

  cheap -repp -cm -default-les=all -packing \
    -nsolutions=1 -mrs -memlimit=4096 -timeout=60 english.grm

best wishes, oe

On Wed, Apr 8, 2015 at 5:53 PM, Xue, Ping <ping.xue at boeing.com> wrote:
> Hi,
> I am looking at the parse trees (and the MRS representations output by the
> PET/ERG system, trying to do some grammar engineering to cover our special
> data. One of the things is that we have to allow “the unknown word
> handling”. We used the command-line option –default-les=all. I wonder what
> the other possible values are for this option. While allowing “unknown word
> handling”,  we hope to constrain it in some way. I would appreciate it if
> you could give me some insights or point me to some documentation. I would
> also appreciate detailed documentation about the notations used in the
> output of the PET/ERG system, namely the parse trees, MRS representations,
> etc generated by the system. I can guess quite a bit but not all.
> Ping
> Ping Xue
> Boeing Research & Technology
> PO Box 3707 MC 7L-43
> Seattle, WA 98124-2207
> (425) 373-2861

More information about the developers mailing list