Fix Ego4D target compatibility in local baselines

run_ego4d_local_baselines.py

@@ -93,6 +93,8 @@ def preflight(args) -> str:
     shape, contract = None, None
     counts = {"visual": [0, 0], "audio": [0, 0]}
     total = 0
+    noncanonical_target_rows = 0
+    target_row_pattern_counts: dict[str, int] = {}
     for index, name in enumerate(names, 1):
         data = load_session_data(
             args.feature_root, args.target_root, name, dataset="ego4d",
@@ -117,6 +119,15 @@ def preflight(args) -> str:
                 counts[modality][1] += len(data.states)
                 offset += width + 1
         total += len(data.states)
+        noncanonical_target_rows += int(
+            data.feature_metadata.get("num_noncanonical_target_rows", 0)
+        )
+        for pattern, count in data.feature_metadata.get(
+            "target_row_pattern_counts", {}
+        ).items():
+            target_row_pattern_counts[pattern] = (
+                target_row_pattern_counts.get(pattern, 0) + int(count)
+            )
         del data  # Validate one clip at a time, including all test clips.
         if not args.no_progress and (index == 1 or index % 25 == 0 or index == len(names)):
             print(f"preflight: {index}/{len(names)} clips", flush=True)
@@ -128,6 +139,19 @@ def preflight(args) -> str:
     if fusion != "single" and len(shape) != 2:
         raise ValueError(f"--layer-fusion {fusion} requires stacked [T,L,D] global states")
     print(f"Validated {total} states; global shape per row={shape}; fusion={fusion}", flush=True)
+    if noncanonical_target_rows:
+        canonical_patterns = {"100", "010", "001", "011"}
+        details = ", ".join(
+            f"[{','.join(pattern)}]={count}"
+            for pattern, count in sorted(target_row_pattern_counts.items())
+            if pattern not in canonical_patterns
+        )
+        print(
+            "Ego4D binary target rows outside the EasyCom four-state "
+            f"constraint: {noncanonical_target_rows}/{total} ({details}); "
+            "preserved exactly for the official LSTR MCE objective.",
+            flush=True,
+        )
     for modality, (valid, count) in counts.items():
         if count:
             note = "; provided boxes, not an online detector" if modality == "visual" else ""