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
docs(examples): example 03 must show the whole usage block it sends readers to
Aashish976880751612
diff --git a/examples/03_reading_the_result.py b/examples/03_reading_the_result.pyindex 22505fb4..0a9073e4 100644--- a/examples/03_reading_the_result.py+++ b/examples/03_reading_the_result.py@@ -13,7 +13,9 @@ banner("03", "Reading the result", """{"model": "laya-rl-agent","answers": {<question id>: <answer>, ...},- "usage": {"input_tokens": <batch total>, "output_tokens": 0}}+ "usage": {"input_tokens": <batch total>, "output_tokens": 0,+ "state_tokens": <the whole state>, "state_tokens_dropped": <worst case>,+ "truncated": <bool>, "truncated_questions": [<question id>, ...]}}`answers` is keyed by the ids you chose, so glue code can index it directly instead ofzipping a list back onto the questions. Each answer carries a `type` matching its@@ -22,7 +24,13 @@ banner("03", "Reading the result", """`usage` is the part readers misread: `input_tokens` is the whole batch, with the statecounted once per question, so it grows with the number of questions rather than being a- context length. `output_tokens` is always 0 -- Laya never generates text.+ context length. `output_tokens` is always 0 -- Laya never generates text. The four keys+ after them are how much of your state the model actually read: `state_tokens` is the whole+ serialized state, `state_tokens_dropped` the most of it any one question's head gave up,+ and `truncated` / `truncated_questions` name the questions that gave something up. A seventh+ key, `options`, appears only when the head budget was tight enough to leave some question's+ options sharing a token span -- that question can no longer tell those options apart, and+ `usage["options"]` is where the call says so.""")agent = load("english")