Aarne Ranta
|
8787b56def
|
option rf -paragraphs
|
2018-09-21 10:40:39 +02:00 |
|
Aarne Ranta
|
013f3573e6
|
added transliteration arabic_unvocalized, which omits the vowels
|
2018-06-12 20:39:39 +02:00 |
|
Aarne Ranta
|
96adbebca8
|
some paradigm extensions
|
2018-01-21 12:18:15 +01:00 |
|
Aarne Ranta
|
79feeaf5b8
|
new option linerize -tabtreebank for tab-separated treebank generation
|
2017-12-06 17:37:12 +01:00 |
|
Aarne Ranta
|
a7e7d8d83c
|
year of 3.9 release corrected
|
2017-10-06 15:38:41 +02:00 |
|
Aarne Ranta
|
8005aee378
|
in Haskell GADT generation, hide Tree in the import of PGF
|
2017-09-06 10:51:43 +02:00 |
|
Aarne Ranta
|
37a1f65f86
|
added aligner.c to Android.mk
|
2017-09-01 09:32:28 +02:00 |
|
Aarne Ranta
|
38fd84a28c
|
Merge branch 'master' of https://github.com/GrammaticalFramework/GF
|
2017-08-21 16:20:27 +03:00 |
|
Aarne Ranta
|
f7312b3f4c
|
ChunkJpn import list updated
|
2017-08-21 16:20:13 +03:00 |
|
Aarne Ranta
|
74fd6eef3b
|
AppUrd added
|
2017-08-21 15:06:46 +03:00 |
|
Aarne Ranta
|
0806b8b80e
|
release notes on RGL and UD
|
2017-08-10 16:53:55 +02:00 |
|
aarne
|
d51cbb0f1a
|
added Arabic question mark to arabic and persian transliterations, as well as the zero-width non-joiner U+200C to persian"
|
2017-06-14 12:32:17 +00:00 |
|
aarne
|
9e206496a7
|
handling combination of &+ and &|
|
2017-06-02 06:11:44 +00:00 |
|
aarne
|
d1eb0aa657
|
handling capitalization (&|) in unlexer -bind
|
2017-06-02 06:06:57 +00:00 |
|
aarne
|
2bdfd27173
|
updated android app help file
|
2017-04-19 17:26:51 +00:00 |
|
aarne
|
7963c85367
|
Translator.java in android app now tries to find the lowercased first word in the lexicon if uppercased fails; added Est and Rus in the help file
|
2017-04-12 14:21:54 +00:00 |
|
aarne
|
f2ba7b64bf
|
commented out case_sensitive=off in App
|
2017-04-12 08:59:58 +00:00 |
|
aarne
|
dd61e1dc19
|
re-enabled Phrasebook in App ; added Russian ; testing with lower-casing the input
|
2017-04-11 15:48:18 +00:00 |
|
aarne
|
b789f90da9
|
separated syncat multiwords in dep tree generation ; added function wildcards to concrete annotatations (see for instance english/LangEng.labels)
|
2017-04-06 15:40:44 +00:00 |
|
aarne
|
ee2f3d085e
|
vd command now reads local concrete configurations to deal with syncat words; TODO: dissolve clustered multiwords bound with +
|
2017-04-06 11:55:21 +00:00 |
|
aarne
|
1e33235144
|
fixed a bug in Eq instance generation for GADT and lexical categories
|
2017-03-22 22:25:21 +00:00 |
|
aarne
|
34627e7427
|
corrected fg generation for lexical categories, to avoid a type error in generated code
|
2017-02-24 20:50:36 +00:00 |
|
aarne
|
65d1867fe3
|
typo in imports of SentencesPes corrected
|
2017-01-30 22:17:00 +00:00 |
|
aarne
|
1dbba8f377
|
added Ice and Nno to synopsis and ResourceDemo, and updated documentation
|
2017-01-19 19:58:48 +00:00 |
|
aarne
|
290ea53ae9
|
news on tech talk, nynorsk, and github mirror problems; contrib github to main meny
|
2016-09-21 12:40:18 +00:00 |
|
aarne
|
8729339d26
|
BNF grammars (.bnf or .cf) can have rules without semicolons, as in the GF book examples. However, if the grammar has any multiline rules, all of them must have semicolons.
|
2016-09-05 21:15:44 +00:00 |
|
aarne
|
a1d226891b
|
rbmt summer school announcement
|
2016-06-26 09:08:30 +00:00 |
|
aarne
|
14ed164794
|
release note on new languages
|
2016-06-16 08:28:54 +00:00 |
|
aarne
|
630b9d7b1e
|
added a default value in dependency visualization for cases using maximum, to cover the case where the list of arcs is empty
|
2016-06-15 10:37:00 +00:00 |
|
aarne
|
7633f53af4
|
release notes on RGL
|
2016-06-10 11:27:24 +00:00 |
|
aarne
|
e755c013f8
|
added PassAgentV2 to App, and the specialized syntax functions to app.probs
|
2016-06-05 17:48:11 +00:00 |
|
aarne
|
e682f0972d
|
made App into a functor to make its body easier to modify; unfortunately, the import list must be copied however
|
2016-06-04 17:07:35 +00:00 |
|
aarne
|
4fea4ace7b
|
vd latex: some adjustments in heights and lengths
|
2016-06-03 16:06:22 +00:00 |
|
aarne
|
f5e2c41dd0
|
vd: arc heights now depend on number of arcs below rather than word distance
|
2016-06-02 18:17:54 +00:00 |
|
aarne
|
a6303fd8d1
|
tokenized apertium text to test with ; RelV2 and QuestV2 in App (Eng and Swe for the moment) with 20% loss of parsing speed but improved quality
|
2016-06-02 06:29:35 +00:00 |
|
aarne
|
74e17aef08
|
made the word length proportionate in dependency visualization, instead of using the max length for every word
|
2016-05-23 19:16:06 +00:00 |
|
aarne
|
808bbf7466
|
enabling V3 and V2V in AppFre
|
2016-05-14 08:55:47 +00:00 |
|
aarne
|
7c8d2f71dc
|
commented out the import of Phrasebook in App with --P; the idea is that application grammar parsing is done in a separate phase, which is faster and more flexible. Also re-enabled ComplV3 and ComplV2V, after testing that they don't reduce the parsing speed too much.
|
2016-05-14 08:30:49 +00:00 |
|
aarne
|
d41d512ae3
|
tweaking Chi language codes to get both speech input and output in Android ; one fix in DictionaryChi
|
2015-12-11 10:56:07 +00:00 |
|
aarne
|
0a38e137b6
|
vd -conll2latex now converts conll to latex. Without going through GF trees, but as a service to the dependency parser community.
|
2015-11-23 10:43:03 +00:00 |
|
aarne
|
1d80db4919
|
AppEst.pgf_c added to the ios project (generated file via xcode)
|
2015-11-19 22:25:18 +00:00 |
|
aarne
|
9335a2997c
|
reduced label size and optimized arc length in vd -output=latex
|
2015-11-18 08:12:26 +00:00 |
|
aarne
|
cda3feaf9f
|
latex visualization of dep trees: explained the program better and eliminated most magic numbers
|
2015-11-17 18:08:32 +00:00 |
|
aarne
|
d6a505169a
|
added -output=latex to visialize_dependencies. This generates more familiar-looking output than the default graphviz, which can moreover be pasted into LaTeX documents. Some more work is needed to make long sentences look nice and fit on a page; a constant word length is now used to simplify computing the coordinates.
|
2015-11-17 09:11:10 +00:00 |
|
aarne
|
b33ea36c18
|
ios fix: spaces should be removed from Chi but not from Cat
|
2015-11-11 15:43:48 +00:00 |
|
aarne
|
d452984bb5
|
src/ui/ios/GF-Offline-Translator tested to work in XCode with the current GF runtime and grammars. However, AppEst crashes the app - remains to find out why
|
2015-11-11 14:19:17 +00:00 |
|
aarne
|
a4919b6c94
|
moved old ios-swift files to the subdirectory src/ui/ios/old-swift
|
2015-11-11 10:56:22 +00:00 |
|
aarne
|
6cb2526253
|
AppEst was missing from darcs, oops
|
2015-11-11 09:40:35 +00:00 |
|
aarne
|
5e4e78097d
|
added Estonian to the Android app
|
2015-11-09 13:32:57 +00:00 |
|
aarne
|
435f04b27f
|
small fixes in Est translation and app.probs
|
2015-11-09 12:44:19 +00:00 |
|
aarne
|
a57eddf1df
|
last fixes to make AppEst compile
|
2015-11-09 11:02:09 +00:00 |
|
aarne
|
73861c2993
|
started Estonian phrasebook modules to test AppEst
|
2015-11-09 10:43:24 +00:00 |
|
aarne
|
ce6557f1f2
|
the visualization commands (aw,vd,vp,vt) can now show multiple trees. Previously they only showed one tree even if there were for instance after ambiguous parsing. The reason was that dot (graphviz) ignored all graphs but the first one. Now the graphs are put into separate files. The 'convert' command from ImageMagick package is used to combine them to one pdf. If this is a problem, the old behaviour can be restored by the -number=1 option to the tree-generating command, which cuts away all trees but the one and doesn't require ImageMagick.
|
2015-11-05 16:47:41 +00:00 |
|
aarne
|
e97fa30c9d
|
vp -showdep: as default label for the first (incl. only) argument, don't use dep#0 but assume it is the head
|
2015-11-05 08:30:13 +00:00 |
|
aarne
|
eb49b6ab56
|
improved documentation of vp -showdep
|
2015-11-05 08:23:33 +00:00 |
|
aarne
|
0786dc6f42
|
dependency labels in parse trees now with the -deps flag, -file=labels_file for configuration. With -nocat option this shows reasonable dep trees, more familiar looking than the vd command. With -showfun flag, the tree gives a rather complete picture of the analysis of the sentence.
|
2015-11-04 20:36:47 +00:00 |
|
aarne
|
e39787ab88
|
prepared visualize_parse for showing dependency labels
|
2015-11-04 17:28:09 +00:00 |
|
aarne
|
a1a602c681
|
rewrote an mkPN application in WordsGer to work around a problem that might be a type checker bug
|
2015-07-23 16:16:38 +00:00 |
|
aarne
|
55e2a4c9cb
|
added PassV2 to App; PassVPSlash did not work as VPSlash was disabled for speed reasons
|
2015-07-21 13:01:29 +00:00 |
|
aarne
|
db23eec58e
|
removed the obsolete examples/translator directory, which caused confusion
|
2015-07-20 14:32:50 +00:00 |
|
aarne
|
3ad6e0c6bc
|
added a space after the tag in word alignment to work around a graphviz bug shown with Chinese characters
|
2015-07-16 08:36:14 +00:00 |
|
aarne
|
5f5db10714
|
updated Android helpfile
|
2015-06-09 09:32:11 +00:00 |
|
aarne
|
4b044566a4
|
note on official vs development version of the Android app
|
2015-06-09 09:28:39 +00:00 |
|
aarne
|
ceba45af5e
|
updated demos/app.html
|
2015-06-09 09:22:59 +00:00 |
|
aarne
|
93bde37a97
|
AppFre.pgf added to repo by mistake, now removed
|
2015-06-02 13:21:55 +00:00 |
|
aarne
|
3643e20406
|
copied examples/app to examples/slowApp; the new app is a bit scaled down but much faster
|
2015-06-02 12:59:29 +00:00 |
|
aarne
|
eb07e4e862
|
French input grammar FreInput separated from Fre
|
2015-06-02 09:45:44 +00:00 |
|
aarne
|
212bfd7acc
|
an input grammar for French - 40 times faster, makes all the difference in the app!
|
2015-06-02 09:13:15 +00:00 |
|
aarne
|
2804160f07
|
some translate.probs decreased ; fine-tuning in pgf-hsbind-trans
|
2015-04-21 14:13:13 +00:00 |
|
aarne
|
0add1bdb20
|
pgf-hsbind-trans now with linearizeAll
|
2015-04-17 14:33:28 +00:00 |
|
aarne
|
a5305f1d9f
|
simplified the example translator and renamed it to pgf-hsbind-trans
|
2015-04-16 13:51:18 +00:00 |
|
aarne
|
2ceeea17fd
|
lower probability for chunked phrases
|
2015-04-16 11:39:28 +00:00 |
|
aarne
|
e2175a31ce
|
restored Cantonese as app Chi speech output: nothing else seems to work
|
2015-04-16 06:52:09 +00:00 |
|
aarne
|
b908b2ba3f
|
extended pgf-shell functionalities: batch translation with many trees, and intervening context-dependent tree manipulation
|
2015-04-15 15:53:44 +00:00 |
|
aarne
|
9aeffa15c9
|
updated app help file ; added backslash to ? in Thai keyboard xml to make it compile
|
2015-04-13 09:45:05 +00:00 |
|
aarne
|
1c1e6bf005
|
Mongolian on index page
|
2015-03-13 15:24:10 +00:00 |
|
aarne
|
8ac46d55f7
|
added Nepali to Setup: it had been missing
|
2015-03-13 15:10:16 +00:00 |
|
aarne
|
0f18e65b5e
|
added Afrikaans to Setup: it had been missing for a while
|
2015-03-13 15:08:03 +00:00 |
|
aarne
|
8fcfc49fed
|
updated Setup and RGL status.html with Mongolian: now it compiles by default
|
2015-03-13 14:51:01 +00:00 |
|
aarne
|
1508fc764d
|
DictionarySwe: kick_V2, mouse_N
|
2015-03-11 08:08:38 +00:00 |
|
aarne
|
4ea4d035dd
|
removing spaces from App output of Chi,Jpn,Tha to get nicer speech output (and of course follow the target lang conventions)
|
2015-02-17 18:11:20 +00:00 |
|
aarne
|
818a119a5b
|
App14 with Tha up and running!
|
2015-02-17 17:18:51 +00:00 |
|
aarne
|
3509ee650d
|
news on Summer School on the main page
|
2015-02-09 09:35:51 +00:00 |
|
aarne
|
340328b8a1
|
Phrasebook compiled for Japanese; weather_N in DictionaryGer
|
2015-01-27 13:18:44 +00:00 |
|
aarne
|
0d8e73847c
|
added Japanese to Android app. Had to change target=android-20 to 21.
|
2015-01-21 22:20:25 +00:00 |
|
aarne
|
6ba8851cc8
|
AppJpn.gf and a Makefile entry for App13 which includes Japanese. Everything compiles now. But Japanese has no support for chunks, because ChunkFunctor does not typecheck for it due to its richer Utt type.
|
2015-01-14 17:22:47 +00:00 |
|
aarne
|
537bc249f9
|
lowered Symb_Chunk probability; fixed translator Makefile
|
2014-12-19 15:17:38 +00:00 |
|
aarne
|
a7b22188c1
|
updates in app help file, e.g. the app name
|
2014-12-11 20:05:57 +00:00 |
|
aarne
|
8ec5910ae2
|
probabilities for new chunk rules ; Symb as another literal in app Makefile
|
2014-12-08 10:20:06 +00:00 |
|
aarne
|
fae9df2517
|
some edits in app help file
|
2014-12-07 17:42:47 +00:00 |
|
aarne
|
82a7e22d2e
|
MissApp.hs: a function for preparing a list of lin rules for missing words
|
2014-12-06 08:39:25 +00:00 |
|
aarne
|
52d3839271
|
a script for analysing missing functions, e.g. "which functions of type Quant are missing in each language"
|
2014-12-05 17:01:58 +00:00 |
|
aarne
|
f600cf35e8
|
fixes in app help file ; commented out fake TTS for Bul and Cat ; remove []_ from output with unlinearized functions, so that speech doesn't get garbled.
|
2014-12-05 09:23:44 +00:00 |
|
aarne
|
db97a2b0e4
|
quick reference text prepended to app help file
|
2014-12-04 20:30:50 +00:00 |
|
aarne
|
8beead40bc
|
Catalan in android app builder ; updated app document with Cat ; substitute voice output for Bulgarian and Catalan ; replaced brackets in unknown words by capitalization (for better speech output ; this is experimental)
|
2014-12-03 14:38:17 +00:00 |
|
aarne
|
7316caf092
|
app Makefile default now App12 with Cat
|
2014-12-03 13:18:48 +00:00 |
|
aarne
|
72e45bbf95
|
Catalan added to App, do 'make App12.pgf'
|
2014-12-03 09:00:59 +00:00 |
|
aarne
|
9ba5e6978b
|
switch PhrasebookBul back to present to avoid cabal build failure. More to investigate here.
|
2014-11-27 17:08:35 +00:00 |
|
aarne
|
1e2627d9af
|
make PhrasebookBul use alltenses to prevent misbehaviour in App
|
2014-11-27 16:36:51 +00:00 |
|
aarne
|
4b075da678
|
a compilable stub of PhrasebookJpn. Sent to a native speaker expert for completions.
|
2014-11-15 09:43:15 +00:00 |
|