General Utilities
This page lists all of Transformers general utility functions that are found in the file utils.py.
Most of those are only useful if you are studying the general code in the library.
Enums and namedtuples
class transformers.utils.ExplicitEnum
< source >( value names = None module = None qualname = None type = None start = 1 )
Enum with more explicit error message for missing values.
class transformers.utils.PaddingStrategy
< source >( value names = None module = None qualname = None type = None start = 1 )
Possible values for the padding argument in PreTrainedTokenizerBase.call(). Useful for tab-completion in an
IDE.
class transformers.TensorType
< source >( value names = None module = None qualname = None type = None start = 1 )
Possible values for the return_tensors argument in PreTrainedTokenizerBase.call(). Useful for
tab-completion in an IDE.
Special Decorators
transformers.utils.add_code_sample_docstrings
< source >( *docstr processor_class = None checkpoint = None output_type = None config_class = None mask = '[MASK]' qa_target_start_index = 14 qa_target_end_index = 15 model_cls = None modality = None expected_output = None expected_loss = None real_checkpoint = None revision = None )
Special Properties
class transformers.utils.cached_property
< source >( fget = None fset = None fdel = None doc = None )
Descriptor that mimics @property but caches output in member variable.
From tensorflow_datasets
Built-in in functools from Python 3.8.
Other Utilities
class transformers.utils._LazyModule
< source >( name: str module_file: str import_structure: typing.Dict[typing.FrozenSet[str], typing.Dict[str, typing.Set[str]]] module_spec: ModuleSpec = None extra_objects: typing.Dict[str, object] = None )
Module class that surfaces all objects but only performs associated imports when the objects are requested.