Explanation of Inference Types in OPF

In the model Parameters, there is a section where we have to specify the inference type. It is written in the example Model Parameters that the following inference types are available.

  • TemporalNextStep: Returns one prediction into the future, assumes input contains temporal semantics.

  • NontemporalClassification: Returns classification of input.

  • TemporalClassification: Returns classification of input, assumes input contains temporal semantics.

  • NontemporalAnomaly: Returns anomaly score.

  • TemporalAnomaly: Returns anomaly score, assumes input contains temporal semantics.

  • NontemporalMultiStep: Returns many prediction into the future.

  • TemporalMultiStep: Returns many prediction into the future, assumes input contains temporal semantics.

2 Likes

I made this a wiki and tried me best to update the page. @scott I’m not sure I captured enough detail, can you have a quick look at this and edit if necessary? Once we have good wording I’ll put it in the docs.

1 Like

Appreciate your help. Thank you so much. But can you please elaborate on what you mean by classification of input?