Available in: GBM, DRF, Deep Learning, Naïve-Bayes
Hit ratios can be used to evaluate the performance of a model. The hit ratio is the percentage of instances where the model correctly predicts the actual class of an observation. The
max_hit_ratio_k option specifies the maximum number of predictions to consider for the hit ratio computation.
Note that this option is available for multiclass problems only and is set to 0 (disabled) by default.