Reference Range Flags

1library(yamaa)23yamaa_spec("spec.yaml")$run()$output

1import yamaa23yamaa.spec("spec.yaml").run().output

1# Reference solution for the yamaa benchmark sdtm-lb-ranges (R track).2#3# Written from the benchmark's full prompt and its inputs alone, as an agent4# would write it. Harbor's oracle agent runs it to check that the prompt can5# be solved and that the grader scores a correct answer 1.67library(dplyr, warn.conflicts = FALSE)8library(readr)910lb_raw <- read_csv(11  "/app/input/lb_raw.csv",12  col_types = cols(LBSEQ = col_integer(), .default = col_character())13)14dm <- read_csv(15  "/app/input/dm.csv",16  col_types = cols(.default = col_character())17)18lbrange <- read_csv(19  "/app/input/lbrange.csv",20  col_types = cols(.default = col_character())21)2223lb <- lb_raw |>24  left_join(dm |> select(USUBJID, SEX), by = "USUBJID") |>25  left_join(lbrange, by = c("LBTESTCD", "SEX")) |>26  mutate(27    DOMAIN = "LB",28    LBSTRESU = UNIT,29    LBSTNRLO = NRLO,30    LBSTNRHI = NRHI,31    LBSTRESN = as.character(LBSTRESN),32    LBNRIND = case_when(33      is.na(LBSTRESN) | LBSTRESN == "" ~ NA_character_,34      is.na(NRLO) | NRLO == "" ~ NA_character_,35      is.na(NRHI) | NRHI == "" ~ NA_character_,36      suppressWarnings(as.numeric(LBSTRESN)) <37        suppressWarnings(as.numeric(NRLO)) ~ "LOW",38      suppressWarnings(as.numeric(LBSTRESN)) >39        suppressWarnings(as.numeric(NRHI)) ~ "HIGH",40      TRUE ~ "NORMAL"41    ),42    LBSTRESN = na_if(LBSTRESN, "")43  ) |>44  arrange(USUBJID, LBSEQ) |>45  select(46    DOMAIN, STUDYID, USUBJID, LBSEQ, LBTESTCD, LBSTRESN, LBSTRESU,47    LBSTNRLO, LBSTNRHI, LBNRIND48  )4950dir.create("/app/output", showWarnings = FALSE)51write_csv(lb, "/app/output/lb.csv", na = "")

1# Reference solution for the yamaa benchmark sdtm-lb-ranges (Python track).2#3# Written from the benchmark's full prompt and its inputs alone, as an agent4# would write it. Harbor's oracle agent runs it to check that the prompt can5# be solved and that the grader scores a correct answer 1.67"""Build the SDTM LB dataset from DM, LBRANGE, and LB_RAW.89One record per collected test result in LB_RAW. The reference range10(unit and limits) is selected by the record's test code together with11the subject's sex from DM. LBNRIND flags the result against the limits.12"""1314import csv15import os1617INPUT_DIR = "/app/input"18OUTPUT_DIR = "/app/output"192021def read_rows(name):22    with open(os.path.join(INPUT_DIR, name), newline="") as handle:23        return list(csv.DictReader(handle))242526def main():27    dm = read_rows("dm.csv")28    lbrange = read_rows("lbrange.csv")29    lb_raw = read_rows("lb_raw.csv")3031    sex_by_subject = {row["USUBJID"]: row["SEX"] for row in dm}32    range_by_key = {33        (row["LBTESTCD"], row["SEX"]): row for row in lbrange34    }3536    records = []37    for row in lb_raw:38        sex = sex_by_subject.get(row["USUBJID"], "")39        entry = range_by_key.get((row["LBTESTCD"], sex))4041        if entry is None:42            unit = ""43            nrlo = ""44            nrhi = ""45        else:46            unit = entry["UNIT"]47            nrlo = entry["NRLO"]48            nrhi = entry["NRHI"]4950        result = row["LBSTRESN"]51        nrind = ""52        if result != "" and nrlo != "" and nrhi != "":53            value = float(result)54            if value < float(nrlo):55                nrind = "LOW"56            elif value > float(nrhi):57                nrind = "HIGH"58            else:59                nrind = "NORMAL"6061        records.append(62            {63                "DOMAIN": "LB",64                "STUDYID": row["STUDYID"],65                "USUBJID": row["USUBJID"],66                "LBSEQ": row["LBSEQ"],67                "LBTESTCD": row["LBTESTCD"],68                "LBSTRESN": result,69                "LBSTRESU": unit,70                "LBSTNRLO": nrlo,71                "LBSTNRHI": nrhi,72                "LBNRIND": nrind,73            }74        )7576    fieldnames = [77        "DOMAIN",78        "STUDYID",79        "USUBJID",80        "LBSEQ",81        "LBTESTCD",82        "LBSTRESN",83        "LBSTRESU",84        "LBSTNRLO",85        "LBSTNRHI",86        "LBNRIND",87    ]88    os.makedirs(OUTPUT_DIR, exist_ok=True)89    with open(90        os.path.join(OUTPUT_DIR, "lb.csv"), "w", newline=""91    ) as handle:92        writer = csv.DictWriter(handle, fieldnames=fieldnames)93        writer.writeheader()94        writer.writerows(records)959697if __name__ == "__main__":98    main()

1/* opensas reference solution for the yamaa benchmark sdtm-lb-ranges.2   Reads the benchmark inputs and derives the requested output.3   Execute with the existing opensas runtime: opensas /app/output/result.sas. */45data dm;6  length STUDYID USUBJID SEX $1024;7  infile "/app/input/dm.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;8  input STUDYID : $char1024. USUBJID : $char1024. SEX : $char1024.;9run;1011data lb_raw;12  length STUDYID USUBJID LBTESTCD $1024;13  infile "/app/input/lb_raw.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;14  input STUDYID : $char1024. USUBJID : $char1024. LBSEQ LBTESTCD : $char1024. LBSTRESN;15run;1617data lbrange;18  length LBTESTCD SEX UNIT $1024;19  infile "/app/input/lbrange.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;20  input LBTESTCD : $char1024. SEX : $char1024. UNIT : $char1024. NRLO NRHI;21run;2223proc sql;24create table joined as select a.*,c.UNIT as LBSTRESU,c.NRLO as LBSTNRLO,c.NRHI as LBSTNRHI from lb_raw a left join dm b on a.STUDYID=b.STUDYID and a.USUBJID=b.USUBJID left join lbrange c on a.LBTESTCD=c.LBTESTCD and b.SEX=c.SEX; quit;25data result; set joined; length DOMAIN $2 LBNRIND $8; DOMAIN='LB'; LBNRIND=''; if not missing(LBSTRESN) and not missing(LBSTNRLO) and not missing(LBSTNRHI) then do; if LBSTRESN<LBSTNRLO then LBNRIND='LOW'; else if LBSTRESN>LBSTNRHI then LBNRIND='HIGH'; else LBNRIND='NORMAL'; end; run;26proc sql;27  create table final as select DOMAIN, STUDYID, USUBJID, LBSEQ, LBTESTCD, LBSTRESN, LBSTRESU, LBSTNRLO, LBSTNRHI, LBNRIND from result;28quit;29proc export data=final outfile="/app/output/lb.csv" dbms=csv replace; run;