fix: filter frozen pools by requested task_types subset

2026-07-16 07:13:31 -04:00
parent e1f08dcd3b
commit ee69721ea3
2 changed files with 34 additions and 1 deletions
@@ -320,9 +320,15 @@ def _filter_untrainable_types(
    """
    diag_by_type = Counter(u.task_type for u in build_units(pools.diagnosis))
    val_by_type = Counter(u.task_type for u in build_units(pools.validation))
+    # 先按调用方显式 task_types 收窄候选集：未请求的题型（哪怕可训练）不得进入
+    # keep，否则冻结全局 pools 后 batch/diagnosis 会训练非请求题型，而 gate 只覆盖
+    # 请求题型 → 静默语义偏差（I-4）。task_types=None 表示全部题型皆为候选。
+    candidates = set(diag_by_type) | set(val_by_type)
+    if task_types is not None:
+        candidates &= set(task_types)
    keep: set[str] = set()
    dropped: list[tuple[str, str]] = []
-    for tt in set(diag_by_type) | set(val_by_type):
+    for tt in candidates:
        n_val = val_by_type.get(tt, 0)
        n_units = diag_by_type.get(tt, 0) + n_val
        if n_val < eval_min_per_class: