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
test(packaging): correct the reason the export list is read from source
Aashishfba584add5ff
diff --git a/tests/test_packaging.py b/tests/test_packaging.pyindex 0f6b0313..ff57baac 100644--- a/tests/test_packaging.py+++ b/tests/test_packaging.py@@ -237,10 +237,10 @@ check("extras/every referenced extra is declared",# README documents wrapping an embedder in it, each of its three shortlisting siblings has a# `:::` directive in reference/helpers.md, and it appeared on no docs page.#-# Only the export list is read, never an attribute, so no lazy name resolves and no checkpoint is-# touched. It is read from the source rather than imported: an editable install registers a-# meta-path finder that wins over the `sys.path.insert` this suite does at the top, so-# `import laya` silently returns *another checkout's* package -- which the mutation check caught.+# Like every other check in this file, this one reads source under the directory it walks and+# imports nothing: no `laya.__getattr__` name resolves, torch never loads, and a computed `__all__`+# returns [] for the guard below to report instead of an import quietly yielding whatever the+# environment happens to have installed.def _exported_names():tree = ast.parse(read(os.path.join("laya", "__init__.py")))for node in tree.body: