Difference between revisions of "Udmurt and Komi"
Jump to navigation
Jump to search
(26 intermediate revisions by 3 users not shown) | |||
Line 1: | Line 1: | ||
+ | |||
+ | The udm-kpv language pair. |
||
+ | |||
+ | ==Monolingual transducers== |
||
+ | |||
+ | ==Installation== |
||
+ | ===Monolingual dependencies=== |
||
+ | <pre> |
||
+ | $ svn co https://victorio.uit.no/langtech/trunk/giella-core |
||
+ | $ cd giella-core |
||
+ | $ ./autogen.sh |
||
+ | $ ./configure |
||
+ | $ make |
||
+ | $ cd .. |
||
+ | </pre> |
||
+ | |||
+ | <pre> |
||
+ | $ svn co https://victorio.uit.no/langtech/trunk/giella-shared |
||
+ | $ cd giella-shared |
||
+ | $ ./autogen.sh |
||
+ | $ ./configure |
||
+ | $ make |
||
+ | $ cd .. |
||
+ | </pre> |
||
+ | |||
+ | Now edit your <tt>$HOME/.bashrc</tt> and add two lines, replacing <tt>/PATH/TO</tt> with the full path to those directories: |
||
+ | |||
+ | <pre> |
||
+ | export GIELLA_CORE=/PATH/TO/giella-core |
||
+ | export GIELLA_SHARED=/PATH/TO/giella-shared |
||
+ | </pre> |
||
+ | |||
+ | And do <tt>source ~/.bashrc</tt> |
||
<pre> |
<pre> |
||
Line 4: | Line 37: | ||
$ cd giella-udm |
$ cd giella-udm |
||
$ ./autogen.sh |
$ ./autogen.sh |
||
− | $ ./configure --enable-apertium |
+ | $ ./configure --with-hfst --enable-reversed-intersect --enable-apertium |
$ make |
$ make |
||
$ cd .. |
$ cd .. |
||
Line 11: | Line 44: | ||
<pre> |
<pre> |
||
$ svn co https://victorio.uit.no/langtech/trunk/langs/kpv giella-kpv |
$ svn co https://victorio.uit.no/langtech/trunk/langs/kpv giella-kpv |
||
− | $ cd giella-kpv |
+ | $ cd giella-kpv |
+ | $ ./autogen.sh |
||
− | $ ./configure --enable-apertium |
||
+ | $ ./configure --with-hfst --enable-reversed-intersect --enable-apertium |
||
$ make |
$ make |
||
$ cd .. |
$ cd .. |
||
</pre> |
</pre> |
||
+ | ===Language pair=== |
||
<pre> |
<pre> |
||
Line 24: | Line 59: | ||
$ make |
$ make |
||
</pre> |
</pre> |
||
+ | |||
+ | ==Google Summer of Code 2018== |
||
+ | Workplan |
||
+ | |||
+ | {|class="wikitable" |
||
+ | ! style="width: 10%" | Week |
||
+ | ! style="width: 15%" | Dates |
||
+ | ! style="width: 35%" | Goals |
||
+ | ! style="width: 10%" | Bidix coverage |
||
+ | ! style="width: 10%" | Udm coverage |
||
+ | ! style="width: 10%" | Komi coverage |
||
+ | |- |
||
+ | ! 1 |
||
+ | | style="text-align:center" | 14.05 - 20.05 |
||
+ | | |
||
+ | * Aligning parallel texts |
||
+ | * Add function words |
||
+ | * Write transfer rules for function words |
||
+ | | style="text-align:center" | |
||
+ | 32.7% |
||
+ | | style="text-align:center" | |
||
+ | 77.4% |
||
+ | | style="text-align:center" | |
||
+ | 96.3% |
||
+ | |- |
||
+ | ! 2 |
||
+ | | style="text-align:center" | 21.05 - 27.05 |
||
+ | | |
||
+ | * Add nominals to bilingual dictionary |
||
+ | | style="text-align:center" | |
||
+ | 51.4% |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | !3 |
||
+ | | style="text-align:center" | 28.05 - 03.06 |
||
+ | | |
||
+ | * Write transfer rules for nominals |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | !4 |
||
+ | | style="text-align:center" | 04.06 - 10.06 |
||
+ | | |
||
+ | * Add verbs to bilingual dictionary |
||
+ | | style="text-align:center" | |
||
+ | 62.4% |
||
+ | | style="text-align:center" | |
||
+ | 86% |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | | style="text-align:center" colspan="6" |'''Midterm evaluation''' |
||
+ | |- |
||
+ | !5 |
||
+ | | style="text-align:center" | 11.06 - 17.06 |
||
+ | | |
||
+ | * Write transfer rules for verbs |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | !6 |
||
+ | | style="text-align:center" | 18.06 - 24.06 |
||
+ | | |
||
+ | * Add stems to Udmurt transducer |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | !7 |
||
+ | | style="text-align:center" | 25.06 - 01.07 |
||
+ | | |
||
+ | * Working on disambiguation |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | !8 |
||
+ | | style="text-align:center" | 02.07 - 08.07 |
||
+ | | |
||
+ | * Add stems to Komi transducer |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | | style="text-align:center" colspan="6" | '''Midterm evaluation''' |
||
+ | |- |
||
+ | !9 |
||
+ | | style="text-align:center" | 09.07 - 15.07 |
||
+ | | |
||
+ | * Working on disambiguation |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | !10-11 |
||
+ | | style="text-align:center" | 16.07 - 29.07 |
||
+ | | |
||
+ | * Testing, debugging |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | !12 |
||
+ | | style="text-align:center" | 30.07 - 05.08 |
||
+ | | |
||
+ | * Cleaning up |
||
+ | * Writing documentation |
||
+ | * Releasing |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | | style="text-align:center" | |
||
+ | |- |
||
+ | | style="text-align:center" colspan="6" | '''Project completed''' |
||
+ | |} |
||
+ | [[Category:Language pairs]] |
Latest revision as of 17:08, 14 June 2018
The udm-kpv language pair.
Contents
Monolingual transducers[edit]
Installation[edit]
Monolingual dependencies[edit]
$ svn co https://victorio.uit.no/langtech/trunk/giella-core $ cd giella-core $ ./autogen.sh $ ./configure $ make $ cd ..
$ svn co https://victorio.uit.no/langtech/trunk/giella-shared $ cd giella-shared $ ./autogen.sh $ ./configure $ make $ cd ..
Now edit your $HOME/.bashrc and add two lines, replacing /PATH/TO with the full path to those directories:
export GIELLA_CORE=/PATH/TO/giella-core export GIELLA_SHARED=/PATH/TO/giella-shared
And do source ~/.bashrc
$ svn co https://victorio.uit.no/langtech/trunk/langs/udm giella-udm $ cd giella-udm $ ./autogen.sh $ ./configure --with-hfst --enable-reversed-intersect --enable-apertium $ make $ cd ..
$ svn co https://victorio.uit.no/langtech/trunk/langs/kpv giella-kpv $ cd giella-kpv $ ./autogen.sh $ ./configure --with-hfst --enable-reversed-intersect --enable-apertium $ make $ cd ..
Language pair[edit]
$ git clone git@github.com:/apertium/apertium-udm-kpv.git $ cd apertium-udm-kpv $ ./autogen.sh --with-lang1=../giella-udm/tools/mt/apertium --with-lang2=../giella-kpv/tools/mt/apertium $ make
Google Summer of Code 2018[edit]
Workplan
Week | Dates | Goals | Bidix coverage | Udm coverage | Komi coverage |
---|---|---|---|---|---|
1 | 14.05 - 20.05 |
|
32.7% |
77.4% |
96.3% |
2 | 21.05 - 27.05 |
|
51.4% |
||
3 | 28.05 - 03.06 |
|
|||
4 | 04.06 - 10.06 |
|
62.4% |
86% |
|
Midterm evaluation | |||||
5 | 11.06 - 17.06 |
|
|||
6 | 18.06 - 24.06 |
|
|||
7 | 25.06 - 01.07 |
|
|||
8 | 02.07 - 08.07 |
|
|||
Midterm evaluation | |||||
9 | 09.07 - 15.07 |
|
|||
10-11 | 16.07 - 29.07 |
|
|||
12 | 30.07 - 05.08 |
|
|||
Project completed |