personalitylinmult.inference.onnx_benchmark
===========================================

.. py:module:: personalitylinmult.inference.onnx_benchmark

.. autoapi-nested-parse::

   Compare a champion's PyTorch and ONNX Runtime paths: speed and accuracy.

   Answers the two questions ONNX export exists to answer: is it actually
   faster, and how much (if any) test-set accuracy does the export cost. Runs
   both paths over the real test split from a built ``fi.h5``, using the exact
   metrics suite training reports (1-MAE, Pearson r, CCC, R2, std_ratio) so the
   comparison is apples-to-apples against the numbers already recorded for the
   champion in README.md's "Pretrained models" table.

   Usage::

       uv run python -m personalitylinmult.inference.onnx_benchmark \\
           --checkpoint results/FI/bigfive/fi_lint_wavlm-ccc-bs256/checkpoints/best.ckpt \\
           --onnx results/FI/bigfive/fi_lint_wavlm-ccc-bs256/checkpoints/best.onnx \\
           --data config/FI/data/wavlm.yaml



Attributes
----------

.. autoapisummary::

   personalitylinmult.inference.onnx_benchmark.logger


Functions
---------

.. autoapisummary::

   personalitylinmult.inference.onnx_benchmark.run_benchmark
   personalitylinmult.inference.onnx_benchmark.main


Module Contents
---------------

.. py:data:: logger

   Module-level logger.

.. py:function:: run_benchmark(checkpoint_path: str, onnx_path: str, data_path: str, n_timing_runs: int = 20) -> dict

   Benchmark a champion's PyTorch vs. ONNX Runtime paths.

   :param checkpoint_path: Path to the trained ``.ckpt``.
   :type checkpoint_path: str
   :param onnx_path: Path to the exported ``.onnx`` (with a
                     ``.onnx.json`` metadata sidecar alongside it).
   :type onnx_path: str
   :param data_path: Path to the data config (e.g.
                     ``config/FI/data/wavlm.yaml``) whose ``fi.h5``/test split this
                     benchmark evaluates against.
   :type data_path: str
   :param n_timing_runs: Forward passes averaged for the per-sample
                         latency numbers (accuracy always runs over the full test split
                         regardless of this value).
   :type n_timing_runs: int

   :returns: ``{"pytorch": {...}, "onnx": {...}}``, each with ``latency_ms``
             and the metrics suite (``mean_one_minus_mae``, ``mean_ccc``, etc.).
   :rtype: dict


.. py:function:: main() -> None

   CLI entry point.