Quickstart

In this section, we assume that you have installed intently-nlu and loaded resources for English:

python -m pip install intently-nlu
python -m intently_nlu download en

The Intently NLU engine needs to be trained on some data before it can start extracting information. Thus, the first thing to do is to build a dataset that can be fed into Intently NLU. For now, we will use this sample dataset which contains data for three intents:

• setTemperature -> "Set the temperature to 20 degrees in the living room"
• turnLightOff -> "Turn on the light in the kitchen"
• turnLightOn -> "Turn off the light in the kitchen"

The format used here is JSON, so let’s load it into a python dict:

import json

with open("path/to/example.json") as f:
    sample_dataset = json.load(f)

Now that we have our dataset, we can move forward to the next step which is building an IntentlyNLUEngine. This is the main object of this lib.

from intently_nlu import Dataset, IntentlyNLUEngine

nlu_engine = IntentlyNLUEngine()

Now that we have our engine object created, we need to feed it with our sample dataset. In general, this action will require some machine learning, so we will actually fit the engine:

nlu_engine.fit(Dataset.from_json(sample_dataset))

Our NLU engine is now trained to recognize new utterances that extend beyond what is strictly contained in the dataset: it is able to generalize.

Let’s try to parse something now!

parsing = nlu_engine.parse_utterance("Set the temperature to 20 degrees in the bedroom!")
print(parsing.as_json())

You should get something that looks like this:

{
  "intent": "smarthome/intents/setTemperature",
  "probability": 1.0,
  "raw_utterance": "Set the temperature to 20 degrees in the bedroom!",
  "resolved_slots": {
    "room": "bedroom",
    "room_temperature": "20 degrees"
  }
}

Congrats, you parsed your first intent!