Pipeline¶

class ionworkspipeline.Pipeline(elements, output_file=None, name=None, description=None)¶

A pipeline is a sequence of pipeline elements, each of which takes a set of parameters as input and returns a set of parameters as output. The output of one element is the input of the next. The pipeline can be run to generate a set of parameters, and a report can be generated to document the pipeline.

Parameters¶

elementsdict of {str: _PipelineElement}: A dictionary of pipeline elements. The name is used to identify the element in the report.
output_filestr, optional: The file to save the parameters to. If None, the parameters are not saved.
namestr, optional: A name for the pipeline. Useful for identification and reporting.
descriptionstr, optional: A description of what the pipeline does.

Extends: ConfigMixin

See also: Pipeline — field-level documentation.

classmethod from_schema(schema)¶

Construct a Pipeline from a validated ionworks.pipeline.Pipeline.

Each entry in schema.elements is dispatched to the matching pipeline element’s from_schema based on the schema class hierarchy: DirectEntry-style schemas, DirectEntryFunctionSchema wrappers, DataFit/ArrayDataFit, Validation, and Calculation are each handled. Pre-built pipeline elements pass through unchanged; raw dicts are forwarded as-is.

Parameters¶

schemaionworks.pipeline.Pipeline: Validated schema describing the pipeline elements and metadata.

Returns¶

Pipeline: Pipeline whose elements are pipeline-side instances ready to run.

run() → ParameterValues¶

Execute all pipeline elements sequentially to generate parameters.

Each element receives the accumulated ParameterValues and returns new parameters to add. Results are combined and optionally saved to a file.

Returns¶

iwp.ParameterValues: Complete set of parameters generated by the pipeline.

Examples¶

Build and run a parameter pipeline:

>>> geom_params = iwp.ParameterValues({
...     "Positive electrode capacity [A.h]": 3.0,
...     "Positive electrode active material volume fraction": 0.65,
...     "Positive electrode thickness [m]": 80e-6,
...     "Electrode area [m2]": 0.1,
... })
>>> pipeline = iwp.Pipeline({
...     "geometry": iwp.direct_entries.DirectEntry(geom_params, "Datasheet"),
...     "capacity": iwp.calculations.ElectrodeCapacity(
...         "positive", use_stoich_window=False
...     ),
... })
>>> result = pipeline.run()
>>> float(result["Maximum concentration in positive electrode [mol.m-3]"]) > 0
True

to_config() → dict¶

Convert the Pipeline back to parser configuration format.

Returns¶

dict: Configuration dictionary that can be passed to parse_pipeline

Every step a pipeline runs is a PipelineElement.

class ionworkspipeline.pipeline.PipelineElement(source: str | None)¶

Base class for pipeline elements. Provides methods for generating a report and latex source code.

Parameters¶

sourcestr: The source of the pipeline element, e.g. a reference to a paper or a description of the method used to generate the parameters.

Extends: OptionsController

run(parameter_values: ParameterValues) → ParameterValues¶

Execute the pipeline element.

Parameters¶

parameter_valuesParameterValues: Input parameters. Must be ParameterValues object.

Returns¶

ParameterValues: New parameters to add to pipeline.

Elements capture their constructor arguments through ConfigMixin, which is what makes to_config() round-trip.

class ionworkspipeline.controllers.ConfigMixin¶

Mixin that automatically captures __init__ parameters for subclasses.

When a class inherits from this mixin, any subclass that defines its own __init__ method will have its parameters automatically captured and stored in self._init_params. This is useful for serialization/deserialization (e.g., to_config methods for reverse parsing).

Additionally, if the __init__ has an “options” parameter, it captures which option keys were explicitly passed in self._options_keys_passed before calling the original __init__.

The captured parameters exclude self and filter out None values.

Example¶

>>> class MyClass(ConfigMixin):
...     def __init__(self, a, b=None, c=10):
...         self.a = a
...         self.c = c
...
>>> obj = MyClass(1, c=20)
>>> obj._init_params
{'a': 1, 'c': 20}

See also: ConfigMixin — field-level documentation.

classmethod from_schema(schema)¶

Construct from a validated ionworks.pipeline instance.

Forwards the schema’s fields as kwargs to cls(...), converting any field that holds a nested schema to its runtime counterpart (detected from the values, not declared). Subclasses whose construction needs more (runtime-only fields, alias remapping, deferred parsing) still override.

classmethod get_init_param_names()¶

Get the list of __init__ parameter names.

Excludes self, *args, and **kwargs.

Returns¶

list[str]: List of parameter names.

class ionworkspipeline.controllers.OptionsController¶: Gives a class an options dict validated against its own defaults.

Common pipelines¶

Collections of pre-defined pipelines for common tasks.

Half-Cell Pipelines