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.
Python calibrationclassificationdecision-modelhuggingfacejevmodernbertmultilingualnlppythonpytorchroutingtyped-decisionszero-shot
tests: pin the encoding on #688's onnx_agent.py read
Nandakishor9cb58b7e2b29
diff --git a/tests/test_onnx_export_batch.py b/tests/test_onnx_export_batch.pyindex 466731ca..2d2e2dcc 100644--- a/tests/test_onnx_export_batch.py+++ b/tests/test_onnx_export_batch.py@@ -202,7 +202,7 @@ check("names/outputs are the literal contract", export_onnx.OUTPUT_NAMES,["logits", "act_logits"])_agent_src = open(os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))),- "laya", "onnx_agent.py")).read()+ "laya", "onnx_agent.py"), encoding="utf-8").read()check_true("names/onnx_agent runs the same output names in the same order",'session.run(["logits", "act_logits"]' in _agent_src, "")for _n in export_onnx.INPUT_NAMES: