Basic Demographics

1library(yamaa)23dm <- yamaa_domain("spec.yaml")$output4dm

1import yamaa23dm = yamaa.yamaa_domain("spec.yaml").output4dm

1# Reference solution for the yamaa benchmark sdtm-dm-basic (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)9library(tidyr)10library(stringr)1112odm <- read_csv(13  "/app/input/odm.csv",14  col_types = cols(.default = col_character())15)1617wide <- odm |>18  pivot_wider(19    id_cols = c(StudyOID, SubjectKey),20    names_from = ItemOID,21    values_from = Value,22    values_fn = first23  )24for (column in c("IT.DM.SEX", "IT.DM.AGE", "IT.DM.ARM")) {25  if (!column %in% names(wide)) wide[[column]] <- NA_character_26}27dm <- wide |>28  mutate(29    DOMAIN = "DM",30    STUDYID = StudyOID,31    USUBJID = SubjectKey,32    SUBJID = SubjectKey,33    # M for Male, F for Female; anything missing, blank, uncollected, or34    # otherwise recorded becomes U.35    SEX = case_when(36      `IT.DM.SEX` == "Male" ~ "M",37      `IT.DM.SEX` == "Female" ~ "F",38      .default = "U"39    ),40    AGE = suppressWarnings(as.integer(`IT.DM.AGE`)),41    ARM = na_if(str_trim(`IT.DM.ARM`), ""),42    # The actual arm always equals the planned arm.43    ACTARM = ARM,44    ARMNRS = if_else(45      is.na(ARM), "Not assigned to treatment arm", NA_character_46    )47  ) |>48  arrange(USUBJID) |>49  select(DOMAIN, STUDYID, USUBJID, SUBJID, SEX, AGE, ARM, ACTARM, ARMNRS)5051dir.create("/app/output", showWarnings = FALSE)52write_csv(dm, "/app/output/dm.csv", na = "")

1# Reference solution for the yamaa benchmark sdtm-dm-basic (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 1.67from pathlib import Path89import polars as pl1011odm = pl.read_csv("/app/input/odm.csv", infer_schema=False)1213wide = odm.pivot(14    index=["StudyOID", "SubjectKey"],15    on="ItemOID",16    values="Value",17    aggregate_function="first",18)1920rows = []21for record in wide.to_dicts():22    sex_raw = record.get("IT.DM.SEX")23    # M for Male, F for Female; anything missing, blank, uncollected, or24    # otherwise recorded becomes U.25    sex = {"Male": "M", "Female": "F"}.get(sex_raw, "U")26    age_raw = (record.get("IT.DM.AGE") or "").strip()27    age = int(float(age_raw)) if age_raw else None28    arm = (record.get("IT.DM.ARM") or "").strip() or None29    # The actual arm always equals the planned arm.30    rows.append(31        {32            "DOMAIN": "DM",33            "STUDYID": record["StudyOID"],34            "USUBJID": record["SubjectKey"],35            "SUBJID": record["SubjectKey"],36            "SEX": sex,37            "AGE": age,38            "ARM": arm,39            "ACTARM": arm,40            "ARMNRS": "Not assigned to treatment arm" if arm is None else None,41        }42    )4344dm = pl.DataFrame(45    rows,46    schema={47        "DOMAIN": pl.String,48        "STUDYID": pl.String,49        "USUBJID": pl.String,50        "SUBJID": pl.String,51        "SEX": pl.String,52        "AGE": pl.Int64,53        "ARM": pl.String,54        "ACTARM": pl.String,55        "ARMNRS": pl.String,56    },57).sort("USUBJID")5859Path("/app/output").mkdir(exist_ok=True)60dm.write_csv("/app/output/dm.csv")

1/* opensas reference solution for the yamaa benchmark sdtm-dm-basic.2   Reads the benchmark inputs and derives the requested output.3   Execute with the existing opensas runtime: opensas /app/output/result.sas. */45data odm;6  length StudyOID MetaDataVersionOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupOID ItemGroupRepeatKey ItemOID Value $1024;7  infile "/app/input/odm.csv" dsd dlm="," firstobs=2 truncover lrecl=32767;8  input StudyOID : $char1024. MetaDataVersionOID : $char1024. SubjectKey : $char1024. StudyEventOID : $char1024. StudyEventRepeatKey : $char1024. ItemGroupOID : $char1024. ItemGroupRepeatKey : $char1024. ItemOID : $char1024. Value : $char1024.;9run;1011proc sort data=odm out=ordered; by StudyOID SubjectKey; run;12data wide;13  length SEX AGE ARM $1024;14  retain SEX AGE ARM;15  set ordered;16  by StudyOID SubjectKey;17  if first.SubjectKey then do;18    SEX=''; AGE=''; ARM='';19  end;20    if ItemOID='IT.DM.SEX' then SEX=Value;21  if ItemOID='IT.DM.AGE' then AGE=Value;22  if ItemOID='IT.DM.ARM' then ARM=Value;23  if last.SubjectKey then output;24run;25data result; set wide; length DOMAIN $2 STUDYID USUBJID SUBJID ACTARM ARMNRS $1024;26DOMAIN='DM'; STUDYID=StudyOID; USUBJID=SubjectKey; SUBJID=SubjectKey;27if SEX='Male' then SEX='M'; else if SEX='Female' then SEX='F'; else SEX='U'; ACTARM=ARM; if missing(ARM) then ARMNRS='Not assigned to treatment arm';28run;29proc sql;30  create table final as select DOMAIN, STUDYID, USUBJID, SUBJID, SEX, AGE, ARM, ACTARM, ARMNRS from result;31quit;32proc export data=final outfile="/app/output/dm.csv" dbms=csv replace; run;