Received this warning while train api using CLU

Question

Have received the below warnings on using the CLU train api

https://learn.microsoft.com/en-us/rest/api/language/2022-10-01-preview/conversational-analysis-authoring/train?tabs=HTTP

Can you please suggest what needs to be done here

"warnings": [
        {
            "code": "Warning",
            "message": "Intent `Read` is tagged in `1` training dataset examples, it is recommended to have `10` tags for better model quality."
        },
        {
            "code": "Warning",
            "message": "Entity `testing` is not tagged in any training dataset examples and was not used in the training input. It is recommended to have `10` tags for better model quality."
        }
    ]

Answer 1

The warnings you received while training with the CLU (Conversational Language Understanding) Train API suggest that you should provide more training examples to improve the quality of your model.

What you can do to address these warnings is as follows:

• For the "Intent Read" warning: There is currently only one example tagged with the "Read" intent in the model.

You should provide more examples to help the model understand this intent.
For better model quality, it is recommended to have at least ten examples for each intent.

Add more training examples that represent the "Read" intent and tag them accordingly to resolve this warning.

• For the "Entity testing" warning: The "testing" entity is not tagged in any of your training dataset examples.

An entity with no examples will not contribute to the model's understanding and may result in poor performance when attempting to recognize that entity in user input.

For better model quality, at least 10 examples should be tagged with each entity.

To resolve this warning, add more training examples that include the "testing" entity and tag them appropriately.

Retrain your model with the CLU Train API after you've added more training examples and tagged them with the appropriate intents and entities.

This should help to improve the quality of your model and resolve the warnings you received during training.