Building a bank currently requires finished attribution scores, because the validate path calls load_scores_loss_signed unconditionally and writes a score_sum column for every row of validation.csv — but when exclude_zero_scores is false, scores are only used for the document count and that reporting column, both of which are avoidable. Making scores optional there would let a filter's random controls be trained during the scoring pass instead of after it, which is about 10 hours on a 1M-document row.
Building a bank currently requires finished attribution scores, because the validate path calls
load_scores_loss_signedunconditionally and writes ascore_sumcolumn for every row ofvalidation.csv— but whenexclude_zero_scoresis false, scores are only used for the document count and that reporting column, both of which are avoidable. Making scores optional there would let a filter's random controls be trained during the scoring pass instead of after it, which is about 10 hours on a 1M-document row.