## summaries.register_tidier()


Register a [tidy](tidy.md#greenwood.tidy) adapter for a model class.


Usage

``` python
summaries.register_tidier(
    class_path,
    fn,
)
```


This is the extension point for adding [tidy()](tidy.md#greenwood.tidy) support to new estimator classes. Each adapter is a callable that accepts a fitted model as its first argument (plus optional keyword arguments like `format=`) and returns a tidy DataFrame.


## Parameters


`class_path: str`  
Fully qualified class name used as the registry key, e.g., `"greenwood._cox.CoxPH"`. Using the string path avoids eager imports.

`fn: Tidier`  
Callable with signature `fn(model, *, format=None, **kwargs) -> DataFrame`.


## Examples

Register a custom tidy adapter for a new estimator class:


``` python
from greenwood.summaries import register_tidier

# Define a tidy adapter that delegates to the model's own output
def _tidy_my_model(model, *, format=None, **kwargs):
    return model.to_frame(format=format)

# Register the adapter for the custom model class
register_tidier("mypackage.MyModel", _tidy_my_model)
```