allennlp.common.registrable

Registrable is a “mixin” for endowing any base class with a named registry for its subclasses and a decorator for registering them.

class allennlp.common.registrable.Registrable

Bases: allennlp.common.from_params.FromParams

Any class that inherits from Registrable gains access to a named registry for its subclasses. To register them, just decorate them with the classmethod @BaseClass.register(name).

After which you can call BaseClass.list_available() to get the keys for the registered subclasses, and BaseClass.by_name(name) to get the corresponding subclass. Note that the registry stores the subclasses themselves; not class instances. In most cases you would then call from_params(params) on the returned subclass.

You can specify a default by setting BaseClass.default_implementation. If it is set, it will be the first element of list_available().

Note that if you use this class to implement a new Registrable abstract class, you must ensure that all subclasses of the abstract class are loaded when the module is loaded, because the subclasses register themselves in their respective files. You can achieve this by having the abstract class and all subclasses in the __init__.py of the module in which they reside (as this causes any import of either the abstract class or a subclass to load all other subclasses and the abstract class).

classmethod by_name(name: str) → Type[~T]

default_implementation: str = None

classmethod list_available() → List[str]

List default first if it exists

classmethod register(name: str, exist_ok=False)

Register a class under a particular name.

Parameters

name: ``str``

The name to register the class under.

exist_ok: ``bool`, optional (default=False)

If True, overwrites any existing models registered under name. Else, throws an error if a model is already registered under name.