topK property
If populated, returns attributions for top K indices of outputs (defaults to 1).
Only applies to Models that predicts more than one outputs (e,g, multi-class Models). When set to -1, returns explanations for all outputs.
Implementation
core.int? topK;