Skip to content

AceOptions

Olga Zamaraeva edited this page Oct 16, 2023 · 17 revisions

Below are the most common command-line usages of ACE, followed by a complete reference for all the command-line options.

Parsing (input is one sentence per line):

ace -g grammar.dat [input-file] [-1 | -n count]

Generating (input is one MRS per line):

ace -g grammar.dat -e [input-file] [-1 | -n count]

Compiling a grammar:

ace -G grammar.dat -g path-to-config.tdl

Complete Reference:

Option Description Task Version
P T G
General options
-V Show ACE version number.
-h Show a short synopsis of usage
-v Increase verbosity; up to 3 -v options can be used if desired. The additional output is unlikely to be useful except for debugging.
--license-info Show ACE's license
Grammar loading/compiling
-g filename Identify the grammar to be used. Normally, a compiled grammar image. When compiling a grammar, the location of that grammar's config.tdl file. 0.9
-G filename Enable grammar-compilation mode, and specify where to save the compiled grammar. 0.9
--gen-qc Generate a quickcheck skeleton based on the unification failures encountered during this run. 0.9.31
Alternative modes
-e Generate, rather than parse. 0.9
-l Enable LUI mode. You must have a working yzlui binary in your PATH.
--input-from-lui Inputs to parse are provided by LUI instead of stdin.
--lui-fd=N LUI front-end is already running and connected to file descriptor N.
--generation-server=fr Operate as the generator in the LOGON architecture: read MRSes from ~/tmp/.transfer.$USER.fr
-E Run the REPP preprocessor on each input and output the result; do not perform token mapping or any further parsing.
-m hostname Join an arbiter grid of parsers.
-t Run as a [incr tsdb()] client.
-O Full-forest parse results; output chart edges instead of derivations.
--itsdb-forest Full-forest parse results with structured output; use with --tsdb-stdout and not with -O 0.9.7
Input/output options
-f Format each EP on its own line when printing MRSes.
-y Enable YY input mode (experimental). 0.9.4
--yy-rules when in YY mode, require input tokens to undergo specific lexical rules 0.9.4
-q Suppress printing the input sentence in cases of successful parsing.
-R Do not output results.
-T Only output MRSes, not trees.
--udx(=all) decorate lexeme nodes with their type; "--udx=all" decorates rule nodes too 0.9.24
--tsdb-stdout produce structured output as expected by tsdb 0.9.24
--tsdb-notes output structured run and parse notes
--report-labels add a tree with the labels (e.g., DT, VP as specified in parse-nodes.tdl in ERG defined in config in the parameter parse-node-labels) 0.9.24
--rooted-derivations include the root in the derivation (if multiple roots apply, the first listed in roots.tdl will be used)
--show-realization-mrses output the final MRS for each realization
--show-realization-trees output the derivation for each realization
--show-probability output the probability from the parse reranker
Processing/filtering/robustness options
-n count Use selective unpacking to enumerate only the top count results.
-1 Equivalent to -n 1.
-X Stop when the first result is found, rather than generating the complete packed forest. Not recommended.
-r roots Supplies a space-delimited list of root instances (overriding the setting compiled into the grammar image).
-L lexemes Disable certain lexemes when parsing.
-p Turn off packing.
--equivalence-packing Disable packing under subsumption; only pack edges that are fully equivalent.
--disable-generalization Turn off subsumption under generalization (useful for debugging) 0.9.17
--disable-subsumption-test Turn off the subsumption check in generation (underspecified generation)
--ubertagging=0.001 enable ubertagging 0.9.21
--pcfg=pcfg use a pcfg approximation to parse (c-saw) and combine its results with the HPSG
Performance
-i Perform a fine-grained timing profile of the operation requested (experimental). 0.9.3
--max-chart-megabytes=X Use up to X megabytes of RAM on chart edges before terminating the forest creation portion of parsing (default 1200).
--max-unpack-megabytes=X Use up to X megabytes of total RAM for chart edges and unpacking operations before cancelling a parse (default 1500).
--max-words=X ignore sentences with more than X words (default 70) 0.9.26
--timeout=X Use up to X seconds of time before cancelling a parse. Off by default.
Undocumented options
-j Reserved for experimental modes.
-s Don't unspecialize extra constraints implied by input EPs.
-F Show all potential PRED values associated to the input tokens by the lexicon and morphology subsystems.
-P Broken.
--packed-edge-ids Show internal edge IDs instead of sanitized ones. Can be useful for debugging.
--show-gen-chart With -l and -e, browse the chart after attempting generation.
--maxent=model.mem Read model.mem and use it for disambiguation.
--tnt-model=/.../path Invoke TNT for tagging, using the given model.
--tnt-max-tags=N Allow TNT to suggest up to N tags per word.
--transfer-config=file.txt Only enable transfer rules whose names are listed in file.txt (one per line).
Clone this wiki locally