Requirements for multiple concepts in v2 predict endpoint

Regular Visitor

I know that there can only be one concept outputted from the brain itself. However I noticed in the v2 docs, it shows that multiple concepts may be returned from requesting predictions (https://learn.microsoft.com/en-us/bonsai/brain-api/v2/predict#response-object).

 

Is this a feature not currently utilized/implemented, or does anyone know what are the requirements to have the actions from multiple concepts returned in the API request?

0 Replies