luis

A Python interface to Microsoft LUIS.


Keywords
nlp, nlu, naturallanguage, text, classification, development
License
MIT
Install
pip install luis==2.0.2

Documentation

https://travis-ci.org/wiseman/pyluis.svg?branch=master

pyluis

This is a Python interface to Microsoft LUIS <https://luis.ai/> (Language Understanding Intelligent Service).

How to use it

After you have created and published your LUIS application, you can use it to analyze text:

>>> import luis

# Use the URL LUIS gives you when you publish your app.
>>> l = luis.Luis(url='https://https://api.projectoxford.ai/luis/...')

# Send text to be analyzed:
>>> r = l.analyze('fly forward 10 feet')

# See all identified intents:
>>> print r.intents
[<Intent intent=u'TRANSLATE' score=1.0>,
 <Intent intent=u'None' score=0.015184097>,
 <Intent intent=u'ROTATE' score=0.00713030435>,
 ...]

# See all identified entities:
>>> print r.entities
[<Entity entity=u'forward' type=u'FORWARD' score=0.9585199 start_index=4 end_index=10>,
 <Entity entity=u'10 feet' type=u'builtin.dimension' score=0.9829198 start_index=12 end_index=18>]

# What was the highest-scored intent?
>>> best = r.best_intent()
>>> print best
<Intent intent=u'TRANSLATE' score=1.0>
>>> print best.intent
u'TRANSLATE'
>>> print best.score
1.0