coco_pipe.diagnostics.streaming_variance
========================================

.. py:module:: coco_pipe.diagnostics.streaming_variance

.. autoapi-nested-parse::

   Exact variance diagnostics over re-iterable feature batches.



Attributes
----------

.. autoapisummary::

   coco_pipe.diagnostics.streaming_variance.FeatureBatch
   coco_pipe.diagnostics.streaming_variance.FeatureBatchSource
   coco_pipe.diagnostics.streaming_variance.DEFAULT_PARTICIPATION_RATIO_MAX_GRAM_BYTES


Functions
---------

.. autoapisummary::

   coco_pipe.diagnostics.streaming_variance.streamed_subject_probe
   coco_pipe.diagnostics.streaming_variance.streamed_variance_decomposition_report


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

.. py:type:: FeatureBatch
   :canonical: tuple[np.ndarray, np.ndarray]


.. py:type:: FeatureBatchSource
   :canonical: Iterable[FeatureBatch] | Callable[[], Iterable[FeatureBatch]]


.. py:data:: DEFAULT_PARTICIPATION_RATIO_MAX_GRAM_BYTES
   :value: 4294967296


.. py:function:: streamed_subject_probe(feature_batches, subject, *, cap = 100, n_splits = 5, seed = 42, blocks=None, epochs = 10, learning_rate = 0.01, weight_decay = 0.0001, _n_features = None, _validated = False)

   Fit a deterministic CPU PyTorch multinomial probe over streamed batches.


.. py:function:: streamed_variance_decomposition_report(feature_batches, subject, label, *, feature_scaling = 'zscore', n_null_permutations = 200, rng = None, probe_blocks=None, probe_cap = 100, probe_n_splits = 5, probe_seed = 42, probe_epochs = 10, probe_learning_rate = 0.01, probe_weight_decay = 0.0001, participation_ratio_max_gram_bytes = DEFAULT_PARTICIPATION_RATIO_MAX_GRAM_BYTES)

   Return the dense diagnostic schema without assembling a feature matrix.

   ``feature_batches`` must be re-iterable (or a callable returning an iterable)
   and yield ``(row_indices, features)`` pairs. Row indices refer to the supplied
   subject/label vectors and must cover them exactly once per complete pass.