In several setups (see below), the MA for verbs is broken: Only a single analysis with vox=a and mod=u is generated, the analyses for passive voice and indicative / subjunctive / jussive mood are missing. (I'm wondering which code base the CALIMA Star analyser web interface at https://calimastar.abudhabi.nyu.edu/analyzer/ is using, because it's displaying six analyses, as expected.)
Example
For the input "يوظف", the following single analysis is returned:
CAMeL Tools v1.4.1 / Python 3.7.14 (copy of the "Guided Tour Colab Notebook", snippet below)
from camel_tools.morphology.database import MorphologyDB
from camel_tools.morphology.analyzer import Analyzer
# First, we need to load a morphological database.
# Here, we load the default database which is used for analyzing
# Modern Standard Arabic.
db = MorphologyDB.builtin_db()
analyzer = Analyzer(db)
analyses = analyzer.analyze('يوظف')
for analysis in analyses:
print(analysis, '\n')
Calima Star uses a database that is not open source unfortunately.
If you have access to SAMA 3.1, I can provide instructions on how to get access to the database.
In several setups (see below), the MA for verbs is broken: Only a single analysis with
vox=a
andmod=u
is generated, the analyses for passive voice and indicative / subjunctive / jussive mood are missing. (I'm wondering which code base the CALIMA Star analyser web interface at https://calimastar.abudhabi.nyu.edu/analyzer/ is using, because it's displaying six analyses, as expected.)Example
For the input "يوظف", the following single analysis is returned:
Test setup
I got this behavior with the following setups: