Flag Measurable Disease at Baseline

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

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

1# Reference solution for the yamaa benchmark adam-adrs-measurable-disease (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)910adsl <- read_csv(11  "/app/input/adsl.csv",12  col_types = cols(.default = col_character())13)14tu <- read_csv(15  "/app/input/tu.csv",16  col_types = cols(TUSEQ = col_integer(), .default = col_character())17)1819# Target disease at screening: a screening tumor identification record20# whose result is target. Later target records do not count.21target_subjects <- tu |>22  filter(23    TUTESTCD %in% "TUMIDENT",24    VISIT %in% "SCREENING",25    TUSTRESC %in% "TARGET"26  ) |>27  distinct(USUBJID)2829adrs <- adsl |>30  mutate(31    PARAMCD = "MDIS",32    PARAM = "Measurable Disease at Baseline",33    AVALC = if_else(USUBJID %in% target_subjects$USUBJID, "Y", "N"),34    AVAL = if_else(AVALC == "Y", 1L, 0L)35  ) |>36  select(STUDYID, USUBJID, PARAMCD, PARAM, AVALC, AVAL) |>37  arrange(USUBJID)3839dir.create("/app/output", showWarnings = FALSE)40write_csv(adrs, "/app/output/adrs.csv", na = "")

1# Reference solution for the yamaa benchmark adam-adrs-measurable-disease (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 1.67import csv89INPUT_DIR = "/app/input"10OUTPUT_DIR = "/app/output"111213def read_csv(path):14    with open(path, newline="", encoding="utf-8") as handle:15        return list(csv.DictReader(handle))161718def main():19    adsl = read_csv(f"{INPUT_DIR}/adsl.csv")20    tu = read_csv(f"{INPUT_DIR}/tu.csv")2122    target_subjects = set()23    for row in tu:24        if (25            row.get("TUTESTCD") == "TUMIDENT"26            and row.get("VISIT") == "SCREENING"27            and (row.get("TUSTRESC") or "").strip() == "TARGET"28        ):29            target_subjects.add((row.get("STUDYID"), row.get("USUBJID")))3031    out_rows = []32    for subject in adsl:33        key = (subject.get("STUDYID"), subject.get("USUBJID"))34        avalc = "Y" if key in target_subjects else "N"35        out_rows.append(36            {37                "STUDYID": subject.get("STUDYID"),38                "USUBJID": subject.get("USUBJID"),39                "PARAMCD": "MDIS",40                "PARAM": "Measurable Disease at Baseline",41                "AVALC": avalc,42                "AVAL": 1 if avalc == "Y" else 0,43            }44        )4546    with open(47        f"{OUTPUT_DIR}/adrs.csv", "w", newline="", encoding="utf-8"48    ) as handle:49        writer = csv.DictWriter(50            handle,51            fieldnames=["STUDYID", "USUBJID", "PARAMCD", "PARAM", "AVALC", "AVAL"],52        )53        writer.writeheader()54        writer.writerows(out_rows)555657if __name__ == "__main__":58    main()

1/* opensas reference solution for the yamaa benchmark adam-adrs-measurable-disease.2   Reads the benchmark inputs and derives the requested output.3   Execute with the existing opensas runtime: opensas /app/output/result.sas. */45data adsl;6  length STUDYID USUBJID $1024;7  infile "/app/input/adsl.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;8  input STUDYID : $char1024. USUBJID : $char1024.;9run;1011data tu;12  length STUDYID USUBJID TUTESTCD VISIT TUSTRESC $1024;13  infile "/app/input/tu.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;14  input STUDYID : $char1024. USUBJID : $char1024. TUSEQ TUTESTCD : $char1024. VISIT : $char1024. TUSTRESC : $char1024.;15run;1617proc sql;18  create table counts as select STUDYID,USUBJID,count(*) as matches from tu where TUTESTCD='TUMIDENT' and VISIT='SCREENING' and TUSTRESC='TARGET' group by STUDYID,USUBJID;19  create table joined as select a.*,b.matches from adsl as a left join counts as b on a.STUDYID=b.STUDYID and a.USUBJID=b.USUBJID;20quit;21data result;22  length PARAMCD $8 PARAM $40 AVALC $1;23  set joined; PARAMCD='MDIS'; PARAM='Measurable Disease at Baseline';24  AVALC='N'; AVAL=0; if matches>0 then do; AVALC='Y'; AVAL=1; end;25run;2627proc sql;28  create table final as select STUDYID, USUBJID, PARAMCD, PARAM, AVALC, AVAL from result;29quit;30proc export data=final outfile="/app/output/adrs.csv" dbms=csv replace; run;