Helper functions for archiving models and restoring archived models.

class allennlp.models.archival.Archive(model, config)[source]

Bases: tuple


Alias for field number 1


Alias for field number 0

allennlp.models.archival.archive_model(serialization_dir: str, weights: str = '') → None[source]

Archives the model weights, its training configuration, and its vocabulary to model.tar.gz


serialization_dir: ``str``

The directory where the weights and vocabulary are written out.

weights: ``str``, optional (default=_DEFAULT_WEIGHTS)

Which weights file to include in the archive. The default is

allennlp.models.archival.load_archive(archive_file: str, cuda_device: int = -1, overrides: str = '') → allennlp.models.archival.Archive[source]

Instantiates an Archive from an archived tar.gz file.


archive_file: ``str``

The archive file to load the model from.

cuda_device: ``int``, optional (default = -1)

If cuda_device is >= 0, the model will be loaded onto the corresponding GPU. Otherwise it will be loaded onto the CPU.