Difference between revisions of "Shallow syntactic function labeller"
Deltamachine (talk | contribs) (Created page with "This is [http://wiki.apertium.org/wiki/User:Deltamachine/proposal Google Summer of Code 2017 project] == Architecture == <p>1. The labeller takes a string in Apertium stream...") |
|||
| (28 intermediate revisions by one other user not shown) | |||
| Line 1: | Line 1: | ||
This is [http://wiki.apertium.org/wiki/User:Deltamachine/proposal Google Summer of Code 2017 project] |
This is [http://wiki.apertium.org/wiki/User:Deltamachine/proposal Google Summer of Code 2017 project] |
||
A repository for the whole project: https://github.com/deltamachine/shallow_syntactic_function_labeller |
|||
== Architecture == |
|||
A workplan and progress notes can be found here: [[Shallow syntactic function labeller/Workplan]] |
|||
<p>1. The labeller takes a string in Apertium stream format with morphological tags:</p> |
|||
== What was done == |
|||
<blockquote> |
|||
1. All needed data for North Sami, Kurmanji, Breton, Kazakh and English was prepared: there are two scripts, one of which creates datasets from UD treebanks (it is able to handle Kurmanji, Breton, Kazakh and English) and the second creates datasets from VISL treebanks (is able to handle North Sami). |
|||
^vino<n><m><sg>$ = INPUT |
|||
</blockquote> |
|||
2. Simple RNN, which is able to label sentences, was built. It works with fastText embeddings for every tag which was seen in the corpus: an embedding for a word is just a sum of all word's tags embeddings. |
|||
<p>2. Parses it into a sequence of morphological tags:</p> |
|||
3. The labeller itself was created. Also the testpack for two language pairs was built: it contains all needed data for sme-nob and kmr-eng, the labeller and installation script. |
|||
<blockquote><n><m><sg> |
|||
</blockquote> |
|||
== List of commits == |
|||
<p>3. Restores the model for this language (which is in the same directory and looks like .json file or like a .pkl file)</p> |
|||
All commits are listed below: |
|||
https://github.com/deltamachine/shallow_syntactic_function_labeller/commits/master |
|||
<p>4. The algorithm analyzes the string and gives a sequence of syntactic tags as an output.</p> |
|||
<blockquote> |
|||
<@nsubj> |
|||
</blockquote> |
|||
<p>5. The labeller applies given labels to the original string:</p> |
|||
<blockquote>^vino<n><m><sg><@nsubj>$ = OUTPUT |
|||
</blockquote> |
|||
<p></p> |
|||
<p> So, in the end there will be a module itself and a file with a model. </p> |
|||
== |
== Description == |
||
The shallow syntactic function labeller takes a string in Apertium stream format, parses it into a sequence of morphological tags and gives it to a classifier. The classifier is a simple RNN model trained on prepared datasets which were made from parsed syntax-labelled corpora (mostly UD-treebanks). The classifier analyzes the given sequence of morphological tags, gives a sequence of labels as an output and the labeller applies these labels to the original string. |
|||
=== Labeller in the pipeline === |
|||
The labeller runs between morphological analyzer or disambiguator and pretransfer. |
|||
For example, in sme-nob it runs between sme-nob-disam and sme-nob-pretransfer, like an original syntax module. |
|||
<pre> |
|||
... | cg-proc 'sme-nob.mor.rlx.bin' | python 'sme-nob-labeller.py' | apertium-pretransfer | lt-proc -b 'sme-nob.autobil.bin' | ... |
|||
</pre> |
|||
=== Language pairs support === |
|||
Currently the labeller works with following language pairs: |
|||
* sme-nob: the labeller may fully replace the original syntax module (it doesn't have all the functionality of the original CG, but works pretty good anyway) |
|||
* kmr-eng: may be tested in the pipeline, but the pair has only a few rules that look at syntax labels |
|||
Also there is all the needed data for Breton, Kazakh and English (https://github.com/deltamachine/shallow_syntactic_function_labeller/tree/master/models), but at this moment br-fr, kk-tat and en-ca just don't have syntax rules, so we can not test the labeller. |
|||
=== Labelling performance === |
|||
The results of validating the labeller on the test set (accuracy = mean accuracy score on the test set). |
|||
{|class=wikitable |
{|class=wikitable |
||
|- |
|- |
||
! |
! Language !! Accuracy |
||
|- |
|- |
||
| |
| North Sami || 81,6% |
||
Handling all the discrepancies between sme-nob tagset and sme-ud.conllu tagset |
|||
|- |
|- |
||
| 2 || 6th June — 12th June || |
|||
|- |
|- |
||
| Kurmanji || 84% |
|||
| 3 || 13th June — 19th June || |
|||
|- |
|||
|- |
|- |
||
| Breton || 79,7% |
|||
| 4 || 20th June — 26th June || |
|||
|- |
|- |
||
! '''First evaluation''' !! colspan="2" align=left | |
|||
Ready-to-use datasets |
|||
|- |
|- |
||
| |
| Kazakh || 82,6% |
||
Building the model |
|||
|- |
|- |
||
| 6 || 4th July — 10th July || |
|||
* Training the classifier |
|||
* Evaluating the quality of the prototype |
|||
|- |
|- |
||
| |
| English || 79,8% |
||
* Further training |
|||
* Working on improvements of the model |
|||
|- |
|||
| 8 || 18th July — 24th July || |
|||
* Final testing |
|||
* Writing a script, which applies labels to the original string in Apertium stream format |
|||
|- |
|||
!'''Second evaluation''' || colspan="2" align=left | |
|||
Well-trained model at least for North Sami |
|||
|- |
|||
| 9 || 25th July — 31th July || |
|||
* Collecting all parts of the labeller together |
|||
* Adding machine-learned module instead of the syntax labelling part of sme-nob CG module to test it |
|||
|- |
|||
| 10 || 1st August — 7th August || |
|||
* Adding machine-learned module instead of the syntax labelling part of sme-nob CG module to test it |
|||
|- |
|||
| 11 || 8th August — 14th August || |
|||
* Testing |
|||
* Fixing bugs |
|||
|- |
|||
| 12 || 15th August — 21th August || |
|||
* Cleaning up the code |
|||
* Writing documentation |
|||
|- |
|||
!'''Final evaluation''' || colspan="2" align=left | |
|||
The prototype shallow syntactic function labeller. |
|||
|- |
|- |
||
|} |
|} |
||
== Installation == |
|||
=== Prerequisites === |
|||
1. Python libraries: |
|||
* DyNet (installation instructions can be found here: http://dynet.readthedocs.io/en/latest/python.html) |
|||
* Streamparser (https://github.com/apertium/streamparser) |
|||
2. Precompiled language pairs which support the labeller (sme-nob, kmr-eng) |
|||
=== How to install a testpack === |
|||
NB: currently the testpack contains syntax modules only for sme-nob and kmr-eng. |
|||
<pre> |
|||
git clone https://github.com/deltamachine/sfl_testpack.git |
|||
cd sfl_testpack |
|||
</pre> |
|||
Script ''setup.py'' adds all the needed files in language pair directory and changes all files with modes. |
|||
'''Arguments:''' |
|||
* ''work_mode:'' '''-lb''' for installing the labeller and changing modes, '''-cg''' for backwarding changes and using the original syntax module (sme-nob.syn.rlx.bin or kmr-eng.prob) in the pipeline. |
|||
* ''lang:'' '''-sme''' for installing/uninstalling the labeller only for sme-nob, '''-kmr''' - only for kmr-eng, '''-all''' - for both. |
|||
For example, this script will install the labeller and add it to the pipeline for both pairs: |
|||
<pre> |
|||
python setup.py -lb -all |
|||
</pre> |
|||
And this script will backward modes changes for sme-nob: |
|||
<pre> |
|||
python setup.py -cg -sme |
|||
</pre> |
|||
== Bugs == |
|||
1. <s>Installation script changes eng-kmr pipeline along with kmr-eng</s> |
|||
2. <s>Problems with tags order (syntactic label is not the last tag)</s> |
|||
3. <s>Words-without-a-label bug</s> |
|||
<pre> |
|||
<spectre> is it possible that some words don't get a label ? |
|||
<spectre> $ echo "Barzanî di peyama xwe de behsa mijarên girîng û kirîtîk kir." | apertium -d . kmr-eng-tagger |
|||
<spectre> ^Barzanî<np><ant><m><sg><obl><@dobj>$ ^di<pr><@case>$ ^peyam<n><f><sg><con><def><@nmod>$ ^xwe<prn><ref><mf><sp><@nmod:poss>$ |
|||
^de<post><@case>$ ^behs<n><f><sg><con><def>$ ^mijar<n><f><pl><con><def><@nmod:poss>$ ^girîng<adj><@amod>$ ^û<cnjcoo><@cc>$ ^*kirîtîk$ |
|||
^kirin<vblex><tv><past><p3><sg>$^..<sent><@punct>$ |
|||
</pre> |
|||
2 and 3 seem to be fixed, but it should be checked carefully. |
|||
== To do == |
|||
* Do more tests. MORE. |
|||
* '''Fix bugs''' |
|||
* Refactore the main code. |
|||
* '''Continue improving the perfomance of the models.''' |
|||
Latest revision as of 01:40, 8 March 2018
This is Google Summer of Code 2017 project
A repository for the whole project: https://github.com/deltamachine/shallow_syntactic_function_labeller
A workplan and progress notes can be found here: Shallow syntactic function labeller/Workplan
Contents
What was done[edit]
1. All needed data for North Sami, Kurmanji, Breton, Kazakh and English was prepared: there are two scripts, one of which creates datasets from UD treebanks (it is able to handle Kurmanji, Breton, Kazakh and English) and the second creates datasets from VISL treebanks (is able to handle North Sami).
2. Simple RNN, which is able to label sentences, was built. It works with fastText embeddings for every tag which was seen in the corpus: an embedding for a word is just a sum of all word's tags embeddings.
3. The labeller itself was created. Also the testpack for two language pairs was built: it contains all needed data for sme-nob and kmr-eng, the labeller and installation script.
List of commits[edit]
All commits are listed below:
https://github.com/deltamachine/shallow_syntactic_function_labeller/commits/master
Description[edit]
The shallow syntactic function labeller takes a string in Apertium stream format, parses it into a sequence of morphological tags and gives it to a classifier. The classifier is a simple RNN model trained on prepared datasets which were made from parsed syntax-labelled corpora (mostly UD-treebanks). The classifier analyzes the given sequence of morphological tags, gives a sequence of labels as an output and the labeller applies these labels to the original string.
Labeller in the pipeline[edit]
The labeller runs between morphological analyzer or disambiguator and pretransfer.
For example, in sme-nob it runs between sme-nob-disam and sme-nob-pretransfer, like an original syntax module.
... | cg-proc 'sme-nob.mor.rlx.bin' | python 'sme-nob-labeller.py' | apertium-pretransfer | lt-proc -b 'sme-nob.autobil.bin' | ...
Language pairs support[edit]
Currently the labeller works with following language pairs:
- sme-nob: the labeller may fully replace the original syntax module (it doesn't have all the functionality of the original CG, but works pretty good anyway)
- kmr-eng: may be tested in the pipeline, but the pair has only a few rules that look at syntax labels
Also there is all the needed data for Breton, Kazakh and English (https://github.com/deltamachine/shallow_syntactic_function_labeller/tree/master/models), but at this moment br-fr, kk-tat and en-ca just don't have syntax rules, so we can not test the labeller.
Labelling performance[edit]
The results of validating the labeller on the test set (accuracy = mean accuracy score on the test set).
| Language | Accuracy |
|---|---|
| North Sami | 81,6% |
| Kurmanji | 84% |
| Breton | 79,7% |
| Kazakh | 82,6% |
| English | 79,8% |
Installation[edit]
Prerequisites[edit]
1. Python libraries:
- DyNet (installation instructions can be found here: http://dynet.readthedocs.io/en/latest/python.html)
- Streamparser (https://github.com/apertium/streamparser)
2. Precompiled language pairs which support the labeller (sme-nob, kmr-eng)
How to install a testpack[edit]
NB: currently the testpack contains syntax modules only for sme-nob and kmr-eng.
git clone https://github.com/deltamachine/sfl_testpack.git cd sfl_testpack
Script setup.py adds all the needed files in language pair directory and changes all files with modes.
Arguments:
- work_mode: -lb for installing the labeller and changing modes, -cg for backwarding changes and using the original syntax module (sme-nob.syn.rlx.bin or kmr-eng.prob) in the pipeline.
- lang: -sme for installing/uninstalling the labeller only for sme-nob, -kmr - only for kmr-eng, -all - for both.
For example, this script will install the labeller and add it to the pipeline for both pairs:
python setup.py -lb -all
And this script will backward modes changes for sme-nob:
python setup.py -cg -sme
Bugs[edit]
1. Installation script changes eng-kmr pipeline along with kmr-eng
2. Problems with tags order (syntactic label is not the last tag)
3. Words-without-a-label bug
<spectre> is it possible that some words don't get a label ? <spectre> $ echo "Barzanî di peyama xwe de behsa mijarên girîng û kirîtîk kir." | apertium -d . kmr-eng-tagger <spectre> ^Barzanî<np><ant><m><sg><obl><@dobj>$ ^di<pr><@case>$ ^peyam<n><f><sg><con><def><@nmod>$ ^xwe<prn><ref><mf><sp><@nmod:poss>$ ^de<post><@case>$ ^behs<n><f><sg><con><def>$ ^mijar<n><f><pl><con><def><@nmod:poss>$ ^girîng<adj><@amod>$ ^û<cnjcoo><@cc>$ ^*kirîtîk$ ^kirin<vblex><tv><past><p3><sg>$^..<sent><@punct>$
2 and 3 seem to be fixed, but it should be checked carefully.
To do[edit]
- Do more tests. MORE.
- Fix bugs
- Refactore the main code.
- Continue improving the perfomance of the models.
