cellmap_flow.finetune.cli
=========================

.. py:module:: cellmap_flow.finetune.cli

.. autoapi-nested-parse::

   The finetune training job's command line: its flags, and what they describe.

   - ``build_arg_parser`` and ``parse_args``: every flag of
     ``python -m cellmap_flow.finetune.finetune_cli``. The job manager builds
     the command, and a running job outlives a dashboard upgrade, so flags only
     ever change compatibly.
   - ``model_config_from_args`` and ``build_target_transform``: the model and
     the training target the flags describe.
   - ``RESTARTABLE_ARGS`` and ``apply_restart_params``: which settings a
     restart may change, and how a restart request changes them.



Attributes
----------

.. autoapisummary::

   cellmap_flow.finetune.cli.logger
   cellmap_flow.finetune.cli.LOSS_TYPES
   cellmap_flow.finetune.cli.OUTPUT_TYPES
   cellmap_flow.finetune.cli.RESTARTABLE_ARGS


Functions
---------

.. autoapisummary::

   cellmap_flow.finetune.cli.build_arg_parser
   cellmap_flow.finetune.cli.parse_args
   cellmap_flow.finetune.cli.build_target_transform
   cellmap_flow.finetune.cli.model_config_from_args
   cellmap_flow.finetune.cli.apply_restart_params
   cellmap_flow.finetune.cli.update_run_metadata


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

.. py:data:: logger

.. py:data:: LOSS_TYPES
   :value: ('dice', 'bce', 'combined', 'mse', 'margin', 'interval', 'flow')


.. py:data:: OUTPUT_TYPES
   :value: ('binary', 'binary_broadcast', 'affinities', 'distance', 'flows')


.. py:function:: build_arg_parser()

.. py:function:: parse_args(argv=None) -> argparse.Namespace

   The job's flags, with the defaults that depend on other flags filled in.


.. py:function:: build_target_transform(args, model_config, output_voxel_size_nm=None)

   Build a TargetTransform based on CLI args.

   ``output_voxel_size_nm`` is the annotation patches' voxel size, which
   the distance bounds of ``--loss-type interval`` are measured in; the
   model's output voxel size when not given.


.. py:function:: model_config_from_args(args) -> cellmap_flow.models.models_config.ModelConfig

   The ModelConfig the command line describes.


.. py:data:: RESTARTABLE_ARGS

.. py:function:: apply_restart_params(args, signal_data: dict)

   Update args with parameters from restart signal and persist to metadata.json.

   :param args: argparse Namespace to update
   :param signal_data: Dict from restart signal file


.. py:function:: update_run_metadata(output_dir, change) -> bool

   Apply ``change(metadata)`` to the run's metadata.json; False if the run has none.

   The dashboard's job monitor updates the same file from its own host,
   within seconds of every restart. The file is read just before it is
   replaced, and replaced whole (write_json_atomically), so neither side
   can read the other's write half done, and the window in which both
   update it at once, and one loses the other's change, is kept short.
   A run that the job manager did not start has no metadata.json, and gets
   none.