Event Findings

1library(yamaa)23fa <- yamaa_domain("spec.yaml")$output4fa

1import yamaa23fa = yamaa.yamaa_domain("spec.yaml").output4fa

1# Reference solution for the yamaa benchmark sdtm-fa-event-findings (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(purrr)1112odm <- read_csv(13  "/app/input/odm.csv",14  col_types = cols(.default = col_character())15)1617long <- odm |>18  mutate(Value = na_if(Value, ""))1920ae <- long |>21  filter(ItemGroupOID == "IG.AE") |>22  select(SubjectKey, StudyOID, StudyEventOID, StudyEventRepeatKey, ItemOID, Value) |>23  pivot_wider(names_from = ItemOID, values_from = Value) |>24  rename(25    AETERM = `IT.AE.AETERM`,26    AELNKID_AE = `IT.AE.AELNKID`27  )2829fa_forms <- long |>30  filter(ItemGroupOID == "IG.FA") |>31  select(32    SubjectKey, StudyOID, StudyEventOID, StudyEventRepeatKey,33    ItemGroupRepeatKey, ItemOID, Value34  ) |>35  pivot_wider(names_from = ItemOID, values_from = Value) |>36  rename(37    LINK_FA = `IT.FA.AELNKID`,38    LOCATION = `IT.FA.LOCATION`,39    SIZE = `IT.FA.SIZE`,40    SIZEU = `IT.FA.SIZEU`,41    BIOPSY = `IT.FA.BIOPSY`,42    FADTC = `IT.FA.FADTC`43  )4445fa <- fa_forms |>46  left_join(47    ae,48    by = c("SubjectKey", "StudyEventOID", "StudyEventRepeatKey")49  ) |>50  filter(!is.na(AETERM) & AETERM != "") |>51  mutate(52    LINK = coalesce(LINK_FA, AELNKID_AE),53    STUDYID = StudyOID.x,54    USUBJID = SubjectKey55  ) |>56  select(STUDYID, USUBJID, LINK, FADTC, SIZEU, AETERM, LOCATION, SIZE, BIOPSY) |>57  pivot_longer(58    cols = c(LOCATION, SIZE, BIOPSY),59    names_to = "WHICH",60    values_to = "FAORRES"61  ) |>62  filter(!is.na(FAORRES)) |>63  mutate(64    ORDER = case_when(65      WHICH == "LOCATION" ~ 0L,66      WHICH == "SIZE" ~ 1L,67      .default = 2L68    ),69    FATESTCD = case_when(70      WHICH == "LOCATION" ~ "LOC",71      WHICH == "SIZE" ~ "SIZE",72      .default = "BIOPSY"73    ),74    FATEST = case_when(75      WHICH == "LOCATION" ~ "Location",76      WHICH == "SIZE" ~ "Size",77      .default = "Biopsied"78    ),79    FAOBJ = AETERM,80    FACAT = "AE",81    IS_SIZE = WHICH == "SIZE",82    FAORRESU = if_else(IS_SIZE, SIZEU, NA_character_),83    FASTRESC = FAORRES,84    FASTRESN = if_else(85      IS_SIZE, suppressWarnings(as.numeric(FAORRES)), NA_real_86    ),87    FASTRESU = if_else(IS_SIZE, SIZEU, NA_character_),88    FALNKID = LINK,89    DOMAIN = "FA",90    TIE = FAORRES91  ) |>92  arrange(STUDYID, USUBJID, LINK, ORDER, TIE) |>93  group_by(STUDYID, USUBJID) |>94  mutate(FASEQ = row_number()) |>95  ungroup() |>96  select(97    DOMAIN, STUDYID, USUBJID, FASEQ, FATESTCD, FATEST, FAOBJ, FACAT,98    FAORRES, FAORRESU, FASTRESC, FASTRESN, FASTRESU, FALNKID, FADTC99  )100101dir.create("/app/output", showWarnings = FALSE)102write_csv(fa, "/app/output/fa.csv", na = "")

1# Reference solution for the yamaa benchmark sdtm-fa-event-findings (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)1213ae_by_occurrence = {}14fa_forms = {}15study_of = {}16for r in odm.to_dicts():17    study_of[r["SubjectKey"]] = r["StudyOID"]18    if r["ItemGroupOID"] == "IG.AE":19        key = (r["SubjectKey"], r["StudyEventOID"], r["StudyEventRepeatKey"])20        ae_by_occurrence.setdefault(key, {})[r["ItemOID"]] = (21            r["Value"] if r["Value"] not in (None, "") else None22        )23    elif r["ItemGroupOID"] == "IG.FA":24        key = (25            r["SubjectKey"],26            r["StudyEventOID"],27            r["StudyEventRepeatKey"],28            r["ItemGroupRepeatKey"],29        )30        fa_forms.setdefault(key, {})[r["ItemOID"]] = (31            r["Value"] if r["Value"] not in (None, "") else None32        )3334test_order = {"IT.FA.LOCATION": 0, "IT.FA.SIZE": 1, "IT.FA.BIOPSY": 2}35test_info = {36    "IT.FA.LOCATION": ("LOC", "Location"),37    "IT.FA.SIZE": ("SIZE", "Size"),38    "IT.FA.BIOPSY": ("BIOPSY", "Biopsied"),39}4041rows = []42for (subj, event, rep, _), form in fa_forms.items():43    ae = ae_by_occurrence.get((subj, event, rep), {})44    aeterm = ae.get("IT.AE.AETERM")45    if not aeterm:46        continue47    link = form.get("IT.FA.AELNKID") or ae.get("IT.AE.AELNKID")48    fadtc = form.get("IT.FA.FADTC")49    sizeu = form.get("IT.FA.SIZEU")50    for oid in ("IT.FA.LOCATION", "IT.FA.SIZE", "IT.FA.BIOPSY"):51        val = form.get(oid)52        if val is None:53            continue54        code, name = test_info[oid]55        is_size = oid == "IT.FA.SIZE"56        try:57            num = float(val) if is_size else None58        except ValueError:59            num = None60        rows.append(61            {62                "DOMAIN": "FA",63                "STUDYID": study_of[subj],64                "USUBJID": subj,65                "LINK": link,66                "ORDER": test_order[oid],67                "TIE": val,68                "FATESTCD": code,69                "FATEST": name,70                "FAOBJ": aeterm,71                "FACAT": "AE",72                "FAORRES": val,73                "FAORRESU": sizeu if is_size else None,74                "FASTRESC": val,75                "FASTRESN": num,76                "FASTRESU": sizeu if is_size else None,77                "FALNKID": link,78                "FADTC": fadtc,79            }80        )8182fa = (83    pl.DataFrame(84        rows,85        schema={86            "DOMAIN": pl.String,87            "STUDYID": pl.String,88            "USUBJID": pl.String,89            "LINK": pl.String,90            "ORDER": pl.Int64,91            "TIE": pl.String,92            "FATESTCD": pl.String,93            "FATEST": pl.String,94            "FAOBJ": pl.String,95            "FACAT": pl.String,96            "FAORRES": pl.String,97            "FAORRESU": pl.String,98            "FASTRESC": pl.String,99            "FASTRESN": pl.Float64,100            "FASTRESU": pl.String,101            "FALNKID": pl.String,102            "FADTC": pl.String,103        },104    )105    .sort(["STUDYID", "USUBJID", "LINK", "ORDER", "TIE"])106    .with_columns(107        pl.int_range(1, pl.len() + 1).over(["STUDYID", "USUBJID"]).alias("FASEQ")108    )109    .select(110        "DOMAIN",111        "STUDYID",112        "USUBJID",113        "FASEQ",114        "FATESTCD",115        "FATEST",116        "FAOBJ",117        "FACAT",118        "FAORRES",119        "FAORRESU",120        "FASTRESC",121        "FASTRESN",122        "FASTRESU",123        "FALNKID",124        "FADTC",125    )126)127128Path("/app/output").mkdir(parents=True, exist_ok=True)129fa.write_csv("/app/output/fa.csv")

1/* opensas reference solution for the yamaa benchmark sdtm-fa-event-findings.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 StudyEventOID StudyEventRepeatKey ItemGroupRepeatKey; run;12data events;13  length FAOBJ AELNKID $1024;14  retain FAOBJ AELNKID;15  set ordered;16  where ItemGroupOID='IG.AE';17  by StudyOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupRepeatKey;18  if first.ItemGroupRepeatKey then do;19    FAOBJ=''; AELNKID='';20  end;21    if ItemOID='IT.AE.AETERM' then FAOBJ=Value;22  if ItemOID='IT.AE.AELNKID' then AELNKID=Value;23  if last.ItemGroupRepeatKey then output;24run;25proc sort data=odm out=ordered; by StudyOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupRepeatKey; run;26data findings;27  length FALNKID LOC SIZE SIZEU BIOPSY FADTC $1024;28  retain FALNKID LOC SIZE SIZEU BIOPSY FADTC;29  set ordered;30  where ItemGroupOID='IG.FA';31  by StudyOID SubjectKey StudyEventOID StudyEventRepeatKey ItemGroupRepeatKey;32  if first.ItemGroupRepeatKey then do;33    FALNKID=''; LOC=''; SIZE=''; SIZEU=''; BIOPSY=''; FADTC='';34  end;35    if ItemOID='IT.FA.AELNKID' then FALNKID=Value;36  if ItemOID='IT.FA.LOCATION' then LOC=Value;37  if ItemOID='IT.FA.SIZE' then SIZE=Value;38  if ItemOID='IT.FA.SIZEU' then SIZEU=Value;39  if ItemOID='IT.FA.BIOPSY' then BIOPSY=Value;40  if ItemOID='IT.FA.FADTC' then FADTC=Value;41  if last.ItemGroupRepeatKey then output;42run;43proc sql;44create table joined as select a.*,b.FAOBJ from findings a inner join events b on a.StudyOID=b.StudyOID and a.SubjectKey=b.SubjectKey and a.FALNKID=b.AELNKID; quit;45data prepared; set joined; length DOMAIN $2 STUDYID USUBJID FATESTCD FATEST FACAT FAORRES FAORRESU FASTRESC FASTRESU $1024;46DOMAIN='FA'; STUDYID=StudyOID; USUBJID=SubjectKey; FACAT='AE'; FAORRESU=''; FASTRESU=''; FASTRESN=.; EVENTREP=input(ItemGroupRepeatKey,best32.);47if not missing(LOC) then do; FATESTCD='LOC'; FATEST='Location'; FAORRES=LOC; FASTRESC=FAORRES; TESTORD=1; output; end;48if not missing(SIZE) then do; FATESTCD='SIZE'; FATEST='Size'; FAORRES=SIZE; FASTRESC=FAORRES; FAORRESU=SIZEU; FASTRESU=SIZEU; FASTRESN=input(SIZE,best32.); TESTORD=2; output; end;49if not missing(BIOPSY) then do; FATESTCD='BIOPSY'; FATEST='Biopsied'; FAORRES=BIOPSY; FASTRESC=FAORRES; FAORRESU=''; FASTRESU=''; FASTRESN=.; TESTORD=3; output; end; run;50proc sort data=prepared; by STUDYID USUBJID FALNKID TESTORD FAORRES; run;51data result; set prepared; by STUDYID USUBJID; retain FASEQ;52  if first.USUBJID then FASEQ=0;53  FASEQ+1;54run;5556proc sql;57  create table final as select DOMAIN, STUDYID, USUBJID, FASEQ, FATESTCD, FATEST, FAOBJ, FACAT, FAORRES, FAORRESU, FASTRESC, FASTRESN, FASTRESU, FALNKID, FADTC from result;58quit;59proc export data=final outfile="/app/output/fa.csv" dbms=csv replace; run;