Hello, I am trying to interpret the probability figure that gets returned by the parse method. In particular, does it reflect only the likelihood of the intent; or does it reflect a joint probability that takes into account the entire parse; including slots. Also wondering if it would be possible to provide some color as to how its calculated; so that users can develop a sense of how reliable the measure is, and whether there are conditions under which it may be unreliable.
Apologies if this is already in the docs; I scoured them but could not find a clear explanation. IMO it would be helpful to discuss this in the Key Concepts section. TIA for any insight!
Question
Hello, I am trying to interpret the probability figure that gets returned by the parse method. In particular, does it reflect only the likelihood of the intent; or does it reflect a joint probability that takes into account the entire parse; including slots. Also wondering if it would be possible to provide some color as to how its calculated; so that users can develop a sense of how reliable the measure is, and whether there are conditions under which it may be unreliable.
Apologies if this is already in the docs; I scoured them but could not find a clear explanation. IMO it would be helpful to discuss this in the Key Concepts section. TIA for any insight!