Class PreTrainedModel

src/transformers/src/transformers/modeling_utils.py:1273–4698

r""" Base class for all models. [`PreTrainedModel`] takes care of storing the configuration of the models and handles methods for loading, downloading and saving models as well as a few methods common to all models to: - resize the input embeddings, - prune heads in the

class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMixin, PeftAdapterMixin):
    r"""
    Base class for all models.

    [`PreTrainedModel`] takes care of storing the configuration of the models and handles methods for loading,
    downloading and saving models as well as a few methods common to all models to:

        - resize the input embeddings,
        - prune heads in the self-attention heads.

    Class attributes (overridden by derived classes):

        - **config_class** ([`PretrainedConfig`]) -- A subclass of [`PretrainedConfig`] to use as configuration class
          for this model architecture.
        - **load_tf_weights** (`Callable`) -- A python *method* for loading a TensorFlow checkpoint in a PyTorch model,
          taking as arguments:

            - **model** ([`PreTrainedModel`]) -- An instance of the model on which to load the TensorFlow checkpoint.
            - **config** ([`PreTrainedConfig`]) -- An instance of the configuration associated to the model.
            - **path** (`str`) -- A path to the TensorFlow checkpoint.

        - **base_model_prefix** (`str`) -- A string indicating the attribute associated to the base model in derived
          classes of the same architecture adding modules on top of the base model.
        - **is_parallelizable** (`bool`) -- A flag indicating whether this model supports model parallelization.
        - **main_input_name** (`str`) -- The name of the principal input to the model (often `input_ids` for NLP
          models, `pixel_values` for vision models and `input_values` for speech models).
    """

    config_class = None
    base_model_prefix = ""
    main_input_name = "input_ids"
    model_tags = None

    _auto_class = None
    _no_split_modules = None
    _skip_keys_device_placement = None
    _keep_in_fp32_modules = None

    # a list of `re` patterns of `state_dict` keys that should be removed from the list of missing
    # keys we find (keys inside the model but not in the checkpoint) and avoid unnecessary warnings.
    _keys_to_ignore_on_load_missing = None
    # a list of `re` patterns of `state_dict` keys that should be removed from the list of
    # unexpected keys we find (keys inside the checkpoint but not the model) and avoid unnecessary
    # warnings.
    _keys_to_ignore_on_load_unexpected = None
    # a list of `state_dict` keys to ignore when saving the model (useful for keys that aren't
    # trained, but which are either deterministic or tied variables)
    _keys_to_ignore_on_save = None
    # a list of `state_dict` keys that are potentially tied to another key in the state_dict.
    _tied_weights_keys = None

    is_parallelizable = False
    supports_gradient_checkpointing = False
    _is_stateful = False

    # Flash Attention 2 support
    _supports_flash_attn_2 = False