a personal assistant
A robust NLU parser that generalizes well to OOV elements and successfully handles core intents.
A regression harness is used to assess coverage.
To update the NLU module run the following command from the nlu directory:
$ python -m rasa_nlu.train --config config.yml --data data/data.md --fixed_model_name model20190119
Use console.py to parse queries:
$ python console.py model20190119 "recommend garage rock from 1991"
Use pytest to assess parser coverage:
$ python -m pytest test_parser.py --disable-pytest-warnings
Supported entity types:
MOD WHO WORK ARTIST LABEL (REGION) (PLACE) TIME FORMAT RELEASE GENRE TEMP_SIG LAB_SIG PLAYLIST
Supported lookups:
(GENRE) (FORMAT) (RELEASE)
Supported intents:
SHARE REMEMBER RECOMMEND PLAYLIST SEARCH PLAY