When we add intents/entities will the model be auto trained?

Soumya Poojari 20 Reputation points
2023-05-12T13:21:35.06+00:00

When we add intents/entities will the model be auto trained?

How is the flow in CLU ? Do we need to call train api post that we need to call deploy api of CLU to train model ?

Can you please briefout here?

Azure AI Language
Azure AI Language
An Azure service that provides natural language capabilities including sentiment analysis, entity extraction, and automated question answering.
358 questions
0 comments No comments
{count} votes

1 answer

Sort by: Most helpful
  1. Sedat SALMAN 13,160 Reputation points
    2023-05-13T18:02:36.26+00:00

    When using Azure's Language Understanding (LUIS) service, after adding or modifying intents and entities, you typically need to train the model manually. Here's the typical flow in LUIS:

    Create or Import Application: You begin by creating a new LUIS application or importing an existing one.

    Create Entities and Intents: Next, you define your intents (what actions you want your application to understand) and entities (the specific details relevant to the intents).

    Add Utterances: For each intent, you add utterances, which are examples of user input. For instance, for a weather bot, you might add utterances like "What's the weather like?" to the "CheckWeather" intent. You can also annotate entities within these utterances.

    Train the Model: Once you've added your intents, entities, and utterances, you need to train the model. This is not an automatic process; you must manually click the "Train" button in the LUIS portal. This step allows LUIS to learn from the examples you've provided.

    Test the Model: After training, you can test the model directly in the LUIS portal. This is an optional but recommended step to ensure that your model is performing as expected.

    Publish the Model: Once you're satisfied with your model's performance, you can publish it. This makes the model available for use in your applications via the LUIS API.

    Use the Model: After the model is published, you can start integrating it into your application using the LUIS runtime API.

    Monitor and Improve: Based on the user interactions, you can continuously monitor and improve your LUIS model.

    So, to answer your question, the training is not automatic. After you modify or add intents/entities, you need to manually train the model. Once the training is done, you need to publish (or deploy) the model for it to be used in your application.

    ref links

    https://learn.microsoft.com/en-us/azure/cognitive-services/LUIS/concepts/entities

    https://learn.microsoft.com/en-us/azure/cognitive-services/LUIS/how-to/entities

    https://learn.microsoft.com/en-us/azure/cognitive-services/LUIS/how-to/train-test

    https://learn.microsoft.com/en-us/azure/cognitive-services/LUIS/how-to/publish