Skip to content

armgit / laya

Public Imported from GitHub
0

Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.

fix(onnx): store the checkpoint identity the calibration mismatch check reads

Aashish82c118ad6b8e
diff --git a/laya/onnx_agent.py b/laya/onnx_agent.py
index 6f6b5f00..693a014e 100644
--- a/laya/onnx_agent.py
+++ b/laya/onnx_agent.py
@@ -99,6 +99,13 @@ class ONNXAgent(HookRegistry):
self._hooks_lock = threading.RLock() if not hooks_concurrent else None
self._hooks_mutex = threading.Lock()
self.model_id = model_id_or_path
+ # The PyTorch `Agent` carries these two names (`agent.py:385-386`) and `calibrate`'s
+ # identity check reads them off whatever object it is handed. Without them here every v2
+ # payload's recorded checkpoint compares against `None`, so the mismatch branch always
+ # fires -- even for the very checkpoint this agent loaded. Mirror `Agent` so the check can
+ # actually match and `load_calibration`'s documented mismatch warning means something.
+ self.model_id_or_path = model_id_or_path
+ self.subfolder = subfolder
import onnxruntime as ort
from transformers import AutoTokenizer
diff --git a/tests/test_calibrate.py b/tests/test_calibrate.py
index acea4547..b2fa5b56 100644
--- a/tests/test_calibrate.py
+++ b/tests/test_calibrate.py
@@ -373,6 +373,41 @@ check_true("onnx/load_calibration documents version/mismatch/clamp behaviour",
check_true("onnx/load_calibration is the shared applier",
"apply_calibration_payload" in inspect.getsource(ONNXAgent.load_calibration))
+# The identity the mismatch check reads must actually be set here, exactly as the PyTorch `Agent`
+# sets it (`agent.py:385-386`). Without both lines every v2 payload compares against `None`, the
+# mismatch branch fires unconditionally on ONNX, and the docstring sentence above is false.
+check_true("onnx/init stores the identity the mismatch check reads",
+ "self.model_id_or_path = model_id_or_path" in onnx_init_src
+ and "self.subfolder = subfolder" in onnx_init_src)
+
+# The behavioural half of that claim, with no weights needed: a stub carrying the same identity
+# `ONNXAgent.__init__` installs. A v2 payload recorded for this very checkpoint loads in silence;
+# one recorded for a different checkpoint warns and still applies. This is the reachability the
+# docstring asserts and the check the four string matches above could not prove.
+onnx_like = type("OnnxLike", (), {})()
+onnx_like.model_id_or_path = "convaiinnovations/laya"
+onnx_like.subfolder = None
+onnx_like.cfg = {"encoder": "answerdotai/ModernBERT-large"}
+matched = calibration_payload(
+ [1.3, 1.2, 1.1], {"choice:2": 1.4},
+ model_id_or_path="convaiinnovations/laya", subfolder=None,
+ config={"encoder": "answerdotai/ModernBERT-large"})
+with warnings.catch_warnings(record=True) as caught:
+ warnings.simplefilter("always")
+ apply_calibration_payload(onnx_like, matched)
+check_true("onnx/matching checkpoint loads without warning", not caught,
+ [str(w.message) for w in caught])
+check("onnx/matching load applies temperature", onnx_like.temperature, [1.3, 1.2, 1.1])
+
+crossed = dict(matched)
+crossed["model_id_or_path"] = "convaiinnovations/laya-multilingual"
+with warnings.catch_warnings(record=True) as caught2:
+ warnings.simplefilter("always")
+ apply_calibration_payload(onnx_like, crossed)
+check_true("onnx/different checkpoint warns", bool(caught2),
+ [str(w.message) for w in caught2])
+check("onnx/different checkpoint still applies", onnx_like.temperature, [1.3, 1.2, 1.1])
+
rec_src = inspect.getsource(records_from_labeled)
check_true("records/_encode_state", "_encode_state" in rec_src)
check_true("records/_decode_answers path", "_decode_answers" in rec_src)