Add exposure cache and keep absolute time only

prepare_data.py

@@ -35,6 +35,17 @@ import pandas as pd  # Pandas for data manipulation
 import tqdm  # Progress bar for chunk processing
 
 
+def _unique_preserve_order(values):
+    """Return unique values while preserving first-seen order."""
+    seen = set()
+    out = []
+    for value in values:
+        if value not in seen:
+            seen.add(value)
+            out.append(value)
+    return out
+
+
 # CSV mapping field IDs to human-readable names
 field_map_file = "field_ids_enriched.csv"
 
@@ -72,7 +83,7 @@ basic_info_fields = [
 
 # Fields needed for tabular extraction from raw CSV.
 tabular_fields = _unique_preserve_order(
-    basic_info_fields + assessment_fields + exposure_fields
+    basic_info_fields + assessment_fields + exposure_fields + ["date_of_birth"]
 )
 
 # TSV mapping field IDs to ICD10-related date columns
@@ -144,6 +155,7 @@ for ukb_chunk in tqdm.tqdm(ukb_iterator, desc="Processing UK Biobank data"):
         ),
         errors="coerce",
     )
+    ukb_chunk["date_of_birth"] = dob.dt.strftime("%Y-%m-%d")
 
     # Use only date variables that actually exist in the current chunk
     present_date_vars = [c for c in date_vars if c in ukb_chunk.columns]
@@ -253,8 +265,11 @@ data = data[np.isin(data[:, 0], valid_eids)]
 final_tabular = final_tabular.loc[valid_eids]
 final_tabular = final_tabular.convert_dtypes()
 
-# Save basic sex information separately.
-basic_info = final_tabular[["sex"]].copy()
+# Save basic information needed by the model and exposure-date alignment.
+basic_cols = ["sex"]
+if "date_of_birth" in final_tabular.columns:
+    basic_cols.append("date_of_birth")
+basic_info = final_tabular[basic_cols].copy()
 basic_info.to_csv("ukb_basic_info.csv")
 
 # Save event data