Deploy Space: update src

src/utils.py

@@ -83,6 +83,19 @@ RESULT_COLUMNS = [
     "num_variates",
 ]
 
+
+def _read_result_csv(path: str | Path) -> pd.DataFrame:
+    df = pd.read_csv(
+        path,
+        usecols=lambda column: column in RESULT_COLUMNS,
+        index_col=False,
+    )
+    missing = [column for column in RESULT_COLUMNS if column not in df.columns]
+    if missing:
+        raise ValueError(f"{path} is missing required result columns: {missing}")
+    return df
+
+
 GIFT_AGGREGATE_FILES = {
     "prediction_length": "results_by_prediction_length.csv",
     "domain": "results_by_domain.csv",
@@ -140,7 +153,7 @@ def load_all_results(root_dir: str = "results") -> pd.DataFrame:
         for file in files:
             if file != "all_results.csv":
                 continue
-            frame = pd.read_csv(os.path.join(subdir, file))
+            frame = _read_result_csv(os.path.join(subdir, file))
             if frame.empty:
                 continue
             frames.append(frame)
@@ -480,6 +493,7 @@ DOMAIN_DISPLAY = {
     "Weather": "Climate",
     "Air Quality": "Climate",
     "Transport": "Transport",
+    "Transportation": "Transportation",
     "Finance": "Finance",
     "Ocean": "Ocean",
     "Hydrology": "Hydrology",
@@ -500,6 +514,9 @@ DATASET_SHORT_NAMES = {
     "usgs_earthquake_all_week_hourly_earthquake_count": "Earthquake count",
     "usgs_potomac_iv_usgs_00060": "Potomac river flow",
     "wikimedia_time_series_hourly_pageviews": "Wikipedia pageviews",
+    "data_gov_sg_singapore_rainfall_5min_rainfall_5min_total": "Singapore rainfall",
+    "data_gov_sg_singapore_taxi_availability_available_taxi_count": "Singapore available taxis",
+    "data_gov_sg_sgx_equities_turnover_monthly_turnover_value_million_sgd": "SGX equities turnover",
 }
 
 
@@ -534,7 +551,7 @@ def lookup_dataset_domain(dataset: str, root_dir: str = "results") -> str:
         if not csv_path.exists():
             continue
         try:
-            df = pd.read_csv(csv_path)
+            df = _read_result_csv(csv_path)
             if "domain" not in df.columns or "dataset" not in df.columns:
                 continue
             # dataset column in all_results is "dataset_id/freq/term", strip suffix
@@ -553,6 +570,7 @@ DOMAIN_TAB_ORDER = [
     "Ocean",
     "Hydrology",
     "Transport",
+    "Transportation",
     "Hazards",
     "Web",
     "Other",
@@ -872,6 +890,7 @@ DOMAIN_COLORS = {
     "Ocean": "#34d399",
     "Hydrology": "#818cf8",
     "Transport": "#fbbf24",
+    "Transportation": "#14b8a6",
     "Hazards": "#fb7185",
     "Web": "#a78bfa",
     "Other": "#94a3b8",