medcat.components.addons.relation_extraction.modernbert.model ============================================================= .. py:module:: medcat.components.addons.relation_extraction.modernbert.model Attributes ---------- .. autoapisummary:: medcat.components.addons.relation_extraction.modernbert.model.logger Classes ------- .. autoapisummary:: medcat.components.addons.relation_extraction.modernbert.model.ConfigRelCAT medcat.components.addons.relation_extraction.modernbert.model.RelExtrBaseModel medcat.components.addons.relation_extraction.modernbert.model.RelExtrModernBertConfig medcat.components.addons.relation_extraction.modernbert.model.RelExtrModernBertModel Functions --------- .. autoapisummary:: medcat.components.addons.relation_extraction.modernbert.model.create_dense_layers Module Contents --------------- .. py:class:: ConfigRelCAT(/, **data) Bases: :py:obj:`medcat.config.config.ComponentConfig` The RelCAT part of the config .. py:attribute:: general :type: General .. py:attribute:: model :type: Model .. py:attribute:: train :type: Train .. py:class:: Config .. py:attribute:: extra :value: 'allow' .. py:attribute:: validate_assignment :value: True .. py:method:: load(load_path = './') :classmethod: Load the config from a file. :param load_path: Path to RelCAT config. Defaults to "./". :type load_path: str :Returns: **ConfigRelCAT** -- The loaded config. .. py:attribute:: comp_name :type: str :value: 'default' The name of the component. If a custom implementation is required, it needs to be registered using `medcat.components.types.register_core_component( , , ) By default, only the 'default' component is registered. .. py:attribute:: _is_dirty :type: bool :value: False .. py:method:: __setattr__(name, value) .. py:property:: is_dirty :type: bool .. py:method:: mark_clean() .. py:method:: get_strategy() .. py:method:: get_init_attrs() :classmethod: .. py:method:: ignore_attrs() :classmethod: .. py:method:: include_properties() :classmethod: .. py:method:: merge_config(other) Merge this config with another config's (partial) model dump. The exepctation is that the `other` dict is a partial model dump. Values specified there are overwritten into the current config. Values not specified there are left intact. The `other` config can have keys/values that do not exist in the config or sub-config. And they will be added where possible. :param other: The model dump :type other: dict :raises IncorrectConfigValues: If unable to set the attribute, trying to set incorrect value, or trying to set sub-config values in an incorrect format (non-dict). .. py:attribute:: model_config :type: ClassVar[pydantic.config.ConfigDict] Configuration for the model, should be a dictionary conforming to [`ConfigDict`][pydantic.config.ConfigDict]. .. py:attribute:: model_fields :type: ClassVar[Dict[str, pydantic.fields.FieldInfo]] Metadata about the fields defined on the model, mapping of field names to [`FieldInfo`][pydantic.fields.FieldInfo] objects. This replaces `Model.__fields__` from Pydantic V1. .. py:attribute:: model_computed_fields :type: ClassVar[Dict[str, pydantic.fields.ComputedFieldInfo]] A dictionary of computed field names and their corresponding `ComputedFieldInfo` objects. .. py:attribute:: __class_vars__ :type: ClassVar[set[str]] The names of the class variables defined on the model. .. py:attribute:: __private_attributes__ :type: ClassVar[Dict[str, pydantic.fields.ModelPrivateAttr]] Metadata about the private attributes of the model. .. py:attribute:: __signature__ :type: ClassVar[inspect.Signature] The synthesized `__init__` [`Signature`][inspect.Signature] of the model. .. py:attribute:: __pydantic_complete__ :type: ClassVar[bool] :value: False Whether model building is completed, or if there are still undefined fields. .. py:attribute:: __pydantic_core_schema__ :type: ClassVar[pydantic_core.CoreSchema] The core schema of the model. .. py:attribute:: __pydantic_custom_init__ :type: ClassVar[bool] Whether the model has a custom `__init__` method. .. py:attribute:: __pydantic_decorators__ :type: ClassVar[pydantic._internal._decorators.DecoratorInfos] Metadata containing the decorators defined on the model. This replaces `Model.__validators__` and `Model.__root_validators__` from Pydantic V1. .. py:attribute:: __pydantic_generic_metadata__ :type: ClassVar[pydantic._internal._generics.PydanticGenericMetadata] Metadata for generic models; contains data used for a similar purpose to __args__, __origin__, __parameters__ in typing-module generics. May eventually be replaced by these. .. py:attribute:: __pydantic_parent_namespace__ :type: ClassVar[Dict[str, Any] | None] :value: None Parent namespace of the model, used for automatic rebuilding of models. .. py:attribute:: __pydantic_post_init__ :type: ClassVar[None | Literal['model_post_init']] The name of the post-init method for the model, if defined. .. py:attribute:: __pydantic_root_model__ :type: ClassVar[bool] :value: False Whether the model is a [`RootModel`][pydantic.root_model.RootModel]. .. py:attribute:: __pydantic_serializer__ :type: ClassVar[pydantic_core.SchemaSerializer] The `pydantic-core` `SchemaSerializer` used to dump instances of the model. .. py:attribute:: __pydantic_validator__ :type: ClassVar[pydantic_core.SchemaValidator | pydantic.plugin._schema_validator.PluggableSchemaValidator] The `pydantic-core` `SchemaValidator` used to validate instances of the model. .. py:attribute:: __pydantic_extra__ :type: dict[str, Any] | None A dictionary containing extra values, if [`extra`][pydantic.config.ConfigDict.extra] is set to `'allow'`. .. py:attribute:: __pydantic_fields_set__ :type: set[str] The names of fields explicitly set during instantiation. .. py:attribute:: __pydantic_private__ :type: dict[str, Any] | None Values of private attributes set on the model instance. .. py:attribute:: __slots__ :value: ('__dict__', '__pydantic_fields_set__', '__pydantic_extra__', '__pydantic_private__') .. py:method:: __init__(/, **data) Create a new model by parsing and validating input data from keyword arguments. Raises [`ValidationError`][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model. `self` is explicitly positional-only to allow `self` as a field name. .. py:property:: model_extra :type: dict[str, Any] | None Get extra fields set during validation. :Returns: **A dictionary of extra fields, or `None` if `config.extra` is not set to `"allow"`.** .. py:property:: model_fields_set :type: set[str] Returns the set of fields that have been explicitly set on this model instance. :Returns: **A set of strings representing the fields that have been set,** -- i.e. that were not filled from defaults. .. py:method:: model_construct(_fields_set = None, **values) :classmethod: Creates a new instance of the `Model` class with validated data. Creates a new model setting `__dict__` and `__pydantic_fields_set__` from trusted or pre-validated data. Default values are respected, but no other validation is performed. !!! note `model_construct()` generally respects the `model_config.extra` setting on the provided model. That is, if `model_config.extra == 'allow'`, then all extra passed values are added to the model instance's `__dict__` and `__pydantic_extra__` fields. If `model_config.extra == 'ignore'` (the default), then all extra passed values are ignored. Because no validation is performed with a call to `model_construct()`, having `model_config.extra == 'forbid'` does not result in an error if extra values are passed, but they will be ignored. :param _fields_set: A set of field names that were originally explicitly set during instantiation. If provided, this is directly used for the [`model_fields_set`][pydantic.BaseModel.model_fields_set] attribute. Otherwise, the field names from the `values` argument will be used. :param values: Trusted or pre-validated data dictionary. :Returns: **A new instance of the `Model` class with validated data.** .. py:method:: model_copy(*, update = None, deep = False) Usage docs: https://docs.pydantic.dev/2.9/concepts/serialization/#model_copy Returns a copy of the model. :param update: Values to change/add in the new model. Note: the data is not validated before creating the new model. You should trust this data. :param deep: Set to `True` to make a deep copy of the model. :Returns: **New model instance.** .. py:method:: model_dump(*, mode = 'python', include = None, exclude = None, context = None, by_alias = False, exclude_unset = False, exclude_defaults = False, exclude_none = False, round_trip = False, warnings = True, serialize_as_any = False) Usage docs: https://docs.pydantic.dev/2.9/concepts/serialization/#modelmodel_dump Generate a dictionary representation of the model, optionally specifying which fields to include or exclude. :param mode: The mode in which `to_python` should run. If mode is 'json', the output will only contain JSON serializable types. If mode is 'python', the output may contain non-JSON-serializable Python objects. :param include: A set of fields to include in the output. :param exclude: A set of fields to exclude from the output. :param context: Additional context to pass to the serializer. :param by_alias: Whether to use the field's alias in the dictionary key if defined. :param exclude_unset: Whether to exclude fields that have not been explicitly set. :param exclude_defaults: Whether to exclude fields that are set to their default value. :param exclude_none: Whether to exclude fields that have a value of `None`. :param round_trip: If True, dumped values should be valid as input for non-idempotent types such as Json[T]. :param warnings: How to handle serialization errors. False/"none" ignores them, True/"warn" logs errors, "error" raises a [`PydanticSerializationError`][pydantic_core.PydanticSerializationError]. :param serialize_as_any: Whether to serialize fields with duck-typing serialization behavior. :Returns: **A dictionary representation of the model.** .. py:method:: model_dump_json(*, indent = None, include = None, exclude = None, context = None, by_alias = False, exclude_unset = False, exclude_defaults = False, exclude_none = False, round_trip = False, warnings = True, serialize_as_any = False) Usage docs: https://docs.pydantic.dev/2.9/concepts/serialization/#modelmodel_dump_json Generates a JSON representation of the model using Pydantic's `to_json` method. :param indent: Indentation to use in the JSON output. If None is passed, the output will be compact. :param include: Field(s) to include in the JSON output. :param exclude: Field(s) to exclude from the JSON output. :param context: Additional context to pass to the serializer. :param by_alias: Whether to serialize using field aliases. :param exclude_unset: Whether to exclude fields that have not been explicitly set. :param exclude_defaults: Whether to exclude fields that are set to their default value. :param exclude_none: Whether to exclude fields that have a value of `None`. :param round_trip: If True, dumped values should be valid as input for non-idempotent types such as Json[T]. :param warnings: How to handle serialization errors. False/"none" ignores them, True/"warn" logs errors, "error" raises a [`PydanticSerializationError`][pydantic_core.PydanticSerializationError]. :param serialize_as_any: Whether to serialize fields with duck-typing serialization behavior. :Returns: **A JSON string representation of the model.** .. py:method:: model_json_schema(by_alias = True, ref_template = DEFAULT_REF_TEMPLATE, schema_generator = GenerateJsonSchema, mode = 'validation') :classmethod: Generates a JSON schema for a model class. :param by_alias: Whether to use attribute aliases or not. :param ref_template: The reference template. :param schema_generator: To override the logic used to generate the JSON schema, as a subclass of `GenerateJsonSchema` with your desired modifications :param mode: The mode in which to generate the schema. :Returns: **The JSON schema for the given model class.** .. py:method:: model_parametrized_name(params) :classmethod: Compute the class name for parametrizations of generic classes. This method can be overridden to achieve a custom naming scheme for generic BaseModels. :param params: Tuple of types of the class. Given a generic class `Model` with 2 type variables and a concrete model `Model[str, int]`, the value `(str, int)` would be passed to `params`. :Returns: **String representing the new class where `params` are passed to `cls` as type variables.** :raises TypeError: Raised when trying to generate concrete names for non-generic models. .. py:method:: model_post_init(__context) Override this method to perform additional initialization after `__init__` and `model_construct`. This is useful if you want to do some validation that requires the entire model to be initialized. .. py:method:: model_rebuild(*, force = False, raise_errors = True, _parent_namespace_depth = 2, _types_namespace = None) :classmethod: Try to rebuild the pydantic-core schema for the model. This may be necessary when one of the annotations is a ForwardRef which could not be resolved during the initial attempt to build the schema, and automatic rebuilding fails. :param force: Whether to force the rebuilding of the model schema, defaults to `False`. :param raise_errors: Whether to raise errors, defaults to `True`. :param _parent_namespace_depth: The depth level of the parent namespace, defaults to 2. :param _types_namespace: The types namespace, defaults to `None`. :Returns: * **Returns `None` if the schema is already "complete" and rebuilding was not required.** * **If rebuilding _was_ required, returns `True` if rebuilding was successful, otherwise `False`.** .. py:method:: model_validate(obj, *, strict = None, from_attributes = None, context = None) :classmethod: Validate a pydantic model instance. :param obj: The object to validate. :param strict: Whether to enforce types strictly. :param from_attributes: Whether to extract data from object attributes. :param context: Additional context to pass to the validator. :raises ValidationError: If the object could not be validated. :Returns: **The validated model instance.** .. py:method:: model_validate_json(json_data, *, strict = None, context = None) :classmethod: Usage docs: https://docs.pydantic.dev/2.9/concepts/json/#json-parsing Validate the given JSON data against the Pydantic model. :param json_data: The JSON data to validate. :param strict: Whether to enforce types strictly. :param context: Extra variables to pass to the validator. :Returns: **The validated Pydantic model.** :raises ValidationError: If `json_data` is not a JSON string or the object could not be validated. .. py:method:: model_validate_strings(obj, *, strict = None, context = None) :classmethod: Validate the given object with string data against the Pydantic model. :param obj: The object containing string data to validate. :param strict: Whether to enforce types strictly. :param context: Extra variables to pass to the validator. :Returns: **The validated Pydantic model.** .. py:method:: __get_pydantic_core_schema__(source, handler, /) :classmethod: Hook into generating the model's CoreSchema. :param source: The class we are generating a schema for. This will generally be the same as the `cls` argument if this is a classmethod. :param handler: A callable that calls into Pydantic's internal CoreSchema generation logic. :Returns: **A `pydantic-core` `CoreSchema`.** .. py:method:: __get_pydantic_json_schema__(core_schema, handler, /) :classmethod: Hook into generating the model's JSON schema. :param core_schema: A `pydantic-core` CoreSchema. You can ignore this argument and call the handler with a new CoreSchema, wrap this CoreSchema (`{'type': 'nullable', 'schema': current_schema}`), or just call the handler with the original schema. :param handler: Call into Pydantic's internal JSON schema generation. This will raise a `pydantic.errors.PydanticInvalidForJsonSchema` if JSON schema generation fails. Since this gets called by `BaseModel.model_json_schema` you can override the `schema_generator` argument to that function to change JSON schema generation globally for a type. :Returns: **A JSON schema, as a Python object.** .. py:method:: __pydantic_init_subclass__(**kwargs) :classmethod: This is intended to behave just like `__init_subclass__`, but is called by `ModelMetaclass` only after the class is actually fully initialized. In particular, attributes like `model_fields` will be present when this is called. This is necessary because `__init_subclass__` will always be called by `type.__new__`, and it would require a prohibitively large refactor to the `ModelMetaclass` to ensure that `type.__new__` was called in such a manner that the class would already be sufficiently initialized. This will receive the same `kwargs` that would be passed to the standard `__init_subclass__`, namely, any kwargs passed to the class definition that aren't used internally by pydantic. :param \*\*kwargs: Any keyword arguments passed to the class definition that aren't used internally by pydantic. .. py:method:: __class_getitem__(typevar_values) :classmethod: .. py:method:: __copy__() Returns a shallow copy of the model. .. py:method:: __deepcopy__(memo = None) Returns a deep copy of the model. .. py:method:: __getattr__(item) .. py:method:: _check_frozen(name, value) .. py:method:: __getstate__() .. py:method:: __setstate__(state) .. py:method:: __eq__(other) .. py:method:: __init_subclass__(**kwargs) :classmethod: This signature is included purely to help type-checkers check arguments to class declaration, which provides a way to conveniently set model_config key/value pairs. ```py from pydantic import BaseModel class MyModel(BaseModel, extra='allow'): ... ``` However, this may be deceiving, since the _actual_ calls to `__init_subclass__` will not receive any of the config arguments, and will only receive any keyword arguments passed during class initialization that are _not_ expected keys in ConfigDict. (This is due to the way `ModelMetaclass.__new__` works.) :param \*\*kwargs: Keyword arguments passed to the class definition, which set model_config .. note:: You may want to override `__pydantic_init_subclass__` instead, which behaves similarly but is called *after* the class is fully initialized. .. py:method:: __iter__() So `dict(model)` works. .. py:method:: __repr__() .. py:method:: __repr_args__() .. py:attribute:: __repr_name__ .. py:attribute:: __repr_str__ .. py:attribute:: __pretty__ .. py:attribute:: __rich_repr__ .. py:method:: __str__() .. py:property:: __fields__ :type: dict[str, pydantic.fields.FieldInfo] .. py:property:: __fields_set__ :type: set[str] .. py:method:: dict(*, include = None, exclude = None, by_alias = False, exclude_unset = False, exclude_defaults = False, exclude_none = False) .. py:method:: json(*, include = None, exclude = None, by_alias = False, exclude_unset = False, exclude_defaults = False, exclude_none = False, encoder = PydanticUndefined, models_as_dict = PydanticUndefined, **dumps_kwargs) .. py:method:: parse_obj(obj) :classmethod: .. py:method:: parse_raw(b, *, content_type = None, encoding = 'utf8', proto = None, allow_pickle = False) :classmethod: .. py:method:: parse_file(path, *, content_type = None, encoding = 'utf8', proto = None, allow_pickle = False) :classmethod: .. py:method:: from_orm(obj) :classmethod: .. py:method:: construct(_fields_set = None, **values) :classmethod: .. py:method:: copy(*, include = None, exclude = None, update = None, deep = False) Returns a copy of the model. !!! warning "Deprecated" This method is now deprecated; use `model_copy` instead. If you need `include` or `exclude`, use: ```py data = self.model_dump(include=include, exclude=exclude, round_trip=True) data = {**data, **(update or {})} copied = self.model_validate(data) ``` :param include: Optional set or mapping specifying which fields to include in the copied model. :param exclude: Optional set or mapping specifying which fields to exclude in the copied model. :param update: Optional dictionary of field-value pairs to override field values in the copied model. :param deep: If True, the values of fields that are Pydantic models will be deep-copied. :Returns: **A copy of the model with included, excluded and updated fields as specified.** .. py:method:: schema(by_alias = True, ref_template = DEFAULT_REF_TEMPLATE) :classmethod: .. py:method:: schema_json(*, by_alias = True, ref_template = DEFAULT_REF_TEMPLATE, **dumps_kwargs) :classmethod: .. py:method:: validate(value) :classmethod: .. py:method:: update_forward_refs(**localns) :classmethod: .. py:method:: _iter(*args, **kwargs) .. py:method:: _copy_and_set_values(*args, **kwargs) .. py:method:: _get_value(*args, **kwargs) :classmethod: .. py:method:: _calculate_keys(*args, **kwargs) .. py:function:: create_dense_layers(relcat_config) .. py:class:: RelExtrBaseModel(relcat_config, model_config, pretrained_model_name_or_path) Bases: :py:obj:`BaseModelBluePrint` Base class for the RelCAT models .. py:attribute:: name :value: 'basemodel_relcat' .. py:method:: __init__(relcat_config, model_config, pretrained_model_name_or_path) Class to hold the HF model + model_config :param pretrained_model_name_or_path: path to load the model from, this can be a HF model i.e: "bert-base-uncased", if left empty, it is normally assumed that a model is loaded from 'model.dat' using the RelCAT.load() method. So if you are initializing/training a model from scratch be sure to base it on some model. :type pretrained_model_name_or_path: str :param relcat_config: relcat config. :type relcat_config: ConfigRelCAT :param model_config: HF bert config for model. :type model_config: PretrainedConfig .. py:attribute:: relcat_config :type: medcat.config.config_rel_cat.ConfigRelCAT .. py:attribute:: model_config :type: medcat.components.addons.relation_extraction.config.RelExtrBaseConfig .. py:attribute:: hf_model .. py:attribute:: pretrained_model_name_or_path :type: str .. py:method:: _reinitialize_dense_and_frozen_layers(relcat_config) Reinitialize the dense layers of the model :param relcat_config: relcat config. :type relcat_config: ConfigRelCAT .. py:method:: forward(input_ids = None, attention_mask = None, token_type_ids = None, position_ids = None, head_mask = None, encoder_hidden_states = None, encoder_attention_mask = None, Q = None, e1_e2_start = None, pooled_output = None) Forward pass for the model :param input_ids: input token ids. Defaults to None. :type input_ids: torch.Tensor :param attention_mask: attention mask for the input ids. Defaults to None. :type attention_mask: torch.Tensor :param token_type_ids: token type ids for the input ids. Defaults to None. :type token_type_ids: torch.Tensor :param position_ids: The position IDs. Defaults to None. :type position_ids: Any :param head_mask: The head mask. Defaults to None. :type head_mask: Any :param encoder_hidden_states: Encoder hidden states. Defaults to None. :type encoder_hidden_states: Any :param encoder_attention_mask: Encoder attention mask. Defaults to None. :type encoder_attention_mask: Any :param Q: Q. Defaults to None. :type Q: Any :param e1_e2_start: Start and end indices for the entities in the input ids. Defaults to None. :type e1_e2_start: Any :param pooled_output: The pooled output. Defaults to None. :type pooled_output: Any :Returns: **Optional[tuple[torch.Tensor, torch.Tensor]]** -- Logits for the relation classification task. .. py:method:: output2logits(pooled_output, sequence_output, input_ids, e1_e2_start) :param pooled_output: embedding of the CLS token :type pooled_output: torch.Tensor :param sequence_output: hidden states/embeddings for each token in the input text :type sequence_output: torch.Tensor :param input_ids: input token ids. :type input_ids: torch.Tensor :param e1_e2_start: annotation tags token position :type e1_e2_start: torch.Tensor :Returns: **torch.Tensor** -- classification probabilities for each token. .. py:method:: load(pretrained_model_name_or_path, relcat_config, model_config) :classmethod: Load the model from the given path :param pretrained_model_name_or_path: path to load the model from. :type pretrained_model_name_or_path: str :param relcat_config: relcat config. :type relcat_config: ConfigRelCAT :param model_config: The model-specific config. :type model_config: RelExtrBaseConfig :returns: **RelExtrBaseModel** -- The loaded model. .. py:attribute:: drop_out :type: torch.nn.Dropout .. py:attribute:: fc1 :type: torch.nn.Linear .. py:attribute:: fc2 :type: torch.nn.Linear .. py:attribute:: fc3 :type: torch.nn.Linear .. py:attribute:: dump_patches :type: bool :value: False .. py:attribute:: _version :type: int :value: 1 This allows better BC support for :meth:`load_state_dict`. In :meth:`state_dict`, the version number will be saved as in the attribute `_metadata` of the returned state dict, and thus pickled. `_metadata` is a dictionary with keys that follow the naming convention of state dict. See ``_load_from_state_dict`` on how to use this information in loading. If new parameters/buffers are added/removed from a module, this number shall be bumped, and the module's `_load_from_state_dict` method can compare the version number and do appropriate changes if the state dict is from before the change. .. py:attribute:: training :type: bool .. py:attribute:: _parameters :type: Dict[str, Optional[torch.nn.parameter.Parameter]] .. py:attribute:: _buffers :type: Dict[str, Optional[torch.Tensor]] .. py:attribute:: _non_persistent_buffers_set :type: Set[str] .. py:attribute:: _backward_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _backward_hooks :type: Dict[int, Callable] .. py:attribute:: _is_full_backward_hook :type: Optional[bool] .. py:attribute:: _forward_hooks :type: Dict[int, Callable] .. py:attribute:: _forward_hooks_with_kwargs :type: Dict[int, bool] .. py:attribute:: _forward_hooks_always_called :type: Dict[int, bool] .. py:attribute:: _forward_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _forward_pre_hooks_with_kwargs :type: Dict[int, bool] .. py:attribute:: _state_dict_hooks :type: Dict[int, Callable] .. py:attribute:: _load_state_dict_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _state_dict_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _load_state_dict_post_hooks :type: Dict[int, Callable] .. py:attribute:: _modules :type: Dict[str, Optional[Module]] .. py:attribute:: call_super_init :type: bool :value: False .. py:attribute:: _compiled_call_impl :type: Optional[Callable] :value: None .. py:method:: register_buffer(name, tensor, persistent = True) Add a buffer to the module. This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm's ``running_mean`` is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting :attr:`persistent` to ``False``. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's :attr:`state_dict`. Buffers can be accessed as attributes using given names. :param name: name of the buffer. The buffer can be accessed from this module using the given name :type name: str :param tensor: buffer to be registered. If ``None``, then operations that run on buffers, such as :attr:`cuda`, are ignored. If ``None``, the buffer is **not** included in the module's :attr:`state_dict`. :type tensor: Tensor or None :param persistent: whether the buffer is part of this module's :attr:`state_dict`. :type persistent: bool Example:: >>> # xdoctest: +SKIP("undefined vars") >>> self.register_buffer('running_mean', torch.zeros(num_features)) .. py:method:: register_parameter(name, param) Add a parameter to the module. The parameter can be accessed as an attribute using given name. :param name: name of the parameter. The parameter can be accessed from this module using the given name :type name: str :param param: parameter to be added to the module. If ``None``, then operations that run on parameters, such as :attr:`cuda`, are ignored. If ``None``, the parameter is **not** included in the module's :attr:`state_dict`. :type param: Parameter or None .. py:method:: add_module(name, module) Add a child module to the current module. The module can be accessed as an attribute using the given name. :param name: name of the child module. The child module can be accessed from this module using the given name :type name: str :param module: child module to be added to the module. :type module: Module .. py:method:: register_module(name, module) Alias for :func:`add_module`. .. py:method:: get_submodule(target) Return the submodule given by ``target`` if it exists, otherwise throw an error. For example, let's say you have an ``nn.Module`` ``A`` that looks like this: .. code-block:: text A( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) ) (The diagram shows an ``nn.Module`` ``A``. ``A`` has a nested submodule ``net_b``, which itself has two submodules ``net_c`` and ``linear``. ``net_c`` then has a submodule ``conv``.) To check whether or not we have the ``linear`` submodule, we would call ``get_submodule("net_b.linear")``. To check whether we have the ``conv`` submodule, we would call ``get_submodule("net_b.net_c.conv")``. The runtime of ``get_submodule`` is bounded by the degree of module nesting in ``target``. A query against ``named_modules`` achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, ``get_submodule`` should always be used. :param target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.) :Returns: **torch.nn.Module** -- The submodule referenced by ``target`` :raises AttributeError: If the target string references an invalid path or resolves to something that is not an ``nn.Module`` .. py:method:: set_submodule(target, module) Set the submodule given by ``target`` if it exists, otherwise throw an error. For example, let's say you have an ``nn.Module`` ``A`` that looks like this: .. code-block:: text A( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) ) (The diagram shows an ``nn.Module`` ``A``. ``A`` has a nested submodule ``net_b``, which itself has two submodules ``net_c`` and ``linear``. ``net_c`` then has a submodule ``conv``.) To overide the ``Conv2d`` with a new submodule ``Linear``, you would call ``set_submodule("net_b.net_c.conv", nn.Linear(33, 16))``. :param target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.) :param module: The module to set the submodule to. :raises ValueError: If the target string is empty :raises AttributeError: If the target string references an invalid path or resolves to something that is not an ``nn.Module`` .. py:method:: get_parameter(target) Return the parameter given by ``target`` if it exists, otherwise throw an error. See the docstring for ``get_submodule`` for a more detailed explanation of this method's functionality as well as how to correctly specify ``target``. :param target: The fully-qualified string name of the Parameter to look for. (See ``get_submodule`` for how to specify a fully-qualified string.) :Returns: **torch.nn.Parameter** -- The Parameter referenced by ``target`` :raises AttributeError: If the target string references an invalid path or resolves to something that is not an ``nn.Parameter`` .. py:method:: get_buffer(target) Return the buffer given by ``target`` if it exists, otherwise throw an error. See the docstring for ``get_submodule`` for a more detailed explanation of this method's functionality as well as how to correctly specify ``target``. :param target: The fully-qualified string name of the buffer to look for. (See ``get_submodule`` for how to specify a fully-qualified string.) :Returns: **torch.Tensor** -- The buffer referenced by ``target`` :raises AttributeError: If the target string references an invalid path or resolves to something that is not a buffer .. py:method:: get_extra_state() Return any extra state to include in the module's state_dict. Implement this and a corresponding :func:`set_extra_state` for your module if you need to store extra state. This function is called when building the module's `state_dict()`. Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes. :Returns: **object** -- Any extra state to store in the module's state_dict .. py:method:: set_extra_state(state) Set extra state contained in the loaded `state_dict`. This function is called from :func:`load_state_dict` to handle any extra state found within the `state_dict`. Implement this function and a corresponding :func:`get_extra_state` for your module if you need to store extra state within its `state_dict`. :param state: Extra state from the `state_dict` :type state: dict .. py:method:: _apply(fn, recurse=True) .. py:method:: apply(fn) Apply ``fn`` recursively to every submodule (as returned by ``.children()``) as well as self. Typical use includes initializing the parameters of a model (see also :ref:`nn-init-doc`). :param fn: function to be applied to each submodule :type fn: :class:`Module` -> None :Returns: **Module** -- self Example:: >>> @torch.no_grad() >>> def init_weights(m): >>> print(m) >>> if type(m) == nn.Linear: >>> m.weight.fill_(1.0) >>> print(m.weight) >>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2)) >>> net.apply(init_weights) Linear(in_features=2, out_features=2, bias=True) Parameter containing: tensor([[1., 1.], [1., 1.]], requires_grad=True) Linear(in_features=2, out_features=2, bias=True) Parameter containing: tensor([[1., 1.], [1., 1.]], requires_grad=True) Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) ) .. py:method:: cuda(device = None) Move all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: ipu(device = None) Move all model parameters and buffers to the IPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: xpu(device = None) Move all model parameters and buffers to the XPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: mtia(device = None) Move all model parameters and buffers to the MTIA. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on MTIA while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: cpu() Move all model parameters and buffers to the CPU. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: type(dst_type) Casts all parameters and buffers to :attr:`dst_type`. .. note:: This method modifies the module in-place. :param dst_type: the desired type :type dst_type: type or string :Returns: **Module** -- self .. py:method:: float() Casts all floating point parameters and buffers to ``float`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: double() Casts all floating point parameters and buffers to ``double`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: half() Casts all floating point parameters and buffers to ``half`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: bfloat16() Casts all floating point parameters and buffers to ``bfloat16`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: to_empty(*, device, recurse = True) Move the parameters and buffers to the specified device without copying storage. :param device: The desired device of the parameters and buffers in this module. :type device: :class:`torch.device` :param recurse: Whether parameters and buffers of submodules should be recursively moved to the specified device. :type recurse: bool :Returns: **Module** -- self .. py:method:: to(device: Optional[torch._prims_common.DeviceLikeType] = ..., dtype: Optional[Module.to.dtype] = ..., non_blocking: bool = ...) -> typing_extensions.Self to(dtype: Module.to.dtype, non_blocking: bool = ...) -> typing_extensions.Self to(tensor: torch.Tensor, non_blocking: bool = ...) -> typing_extensions.Self Move and/or cast the parameters and buffers. This can be called as .. function:: to(device=None, dtype=None, non_blocking=False) :noindex: .. function:: to(dtype, non_blocking=False) :noindex: .. function:: to(tensor, non_blocking=False) :noindex: .. function:: to(memory_format=torch.channels_last) :noindex: Its signature is similar to :meth:`torch.Tensor.to`, but only accepts floating point or complex :attr:`dtype`\ s. In addition, this method will only cast the floating point or complex parameters and buffers to :attr:`dtype` (if given). The integral parameters and buffers will be moved :attr:`device`, if that is given, but with dtypes unchanged. When :attr:`non_blocking` is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices. See below for examples. .. note:: This method modifies the module in-place. :param device: the desired device of the parameters and buffers in this module :type device: :class:`torch.device` :param dtype: the desired floating point or complex dtype of the parameters and buffers in this module :type dtype: :class:`torch.dtype` :param tensor: Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module :type tensor: torch.Tensor :param memory_format: the desired memory format for 4D parameters and buffers in this module (keyword only argument) :type memory_format: :class:`torch.memory_format` :Returns: **Module** -- self Examples:: >>> # xdoctest: +IGNORE_WANT("non-deterministic") >>> linear = nn.Linear(2, 2) >>> linear.weight Parameter containing: tensor([[ 0.1913, -0.3420], [-0.5113, -0.2325]]) >>> linear.to(torch.double) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1913, -0.3420], [-0.5113, -0.2325]], dtype=torch.float64) >>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_CUDA1) >>> gpu1 = torch.device("cuda:1") >>> linear.to(gpu1, dtype=torch.half, non_blocking=True) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1914, -0.3420], [-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1') >>> cpu = torch.device("cpu") >>> linear.to(cpu) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1914, -0.3420], [-0.5112, -0.2324]], dtype=torch.float16) >>> linear = nn.Linear(2, 2, bias=None).to(torch.cdouble) >>> linear.weight Parameter containing: tensor([[ 0.3741+0.j, 0.2382+0.j], [ 0.5593+0.j, -0.4443+0.j]], dtype=torch.complex128) >>> linear(torch.ones(3, 2, dtype=torch.cdouble)) tensor([[0.6122+0.j, 0.1150+0.j], [0.6122+0.j, 0.1150+0.j], [0.6122+0.j, 0.1150+0.j]], dtype=torch.complex128) .. py:method:: register_full_backward_pre_hook(hook, prepend = False) Register a backward pre-hook on the module. The hook will be called every time the gradients for the module are computed. The hook should have the following signature:: hook(module, grad_output) -> tuple[Tensor] or None The :attr:`grad_output` is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of :attr:`grad_output` in subsequent computations. Entries in :attr:`grad_output` will be ``None`` for all non-Tensor arguments. For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function. .. warning :: Modifying inputs inplace is not allowed when using backward hooks and will raise an error. :param hook: The user-defined hook to be registered. :type hook: Callable :param prepend: If true, the provided ``hook`` will be fired before all existing ``backward_pre`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``backward_pre`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``backward_pre`` hooks registered with :func:`register_module_full_backward_pre_hook` will fire before all hooks registered by this method. :type prepend: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: register_backward_hook(hook) Register a backward hook on the module. This function is deprecated in favor of :meth:`~torch.nn.Module.register_full_backward_hook` and the behavior of this function will change in future versions. :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: register_full_backward_hook(hook, prepend = False) Register a backward hook on the module. The hook will be called every time the gradients with respect to a module are computed, i.e. the hook will execute if and only if the gradients with respect to module outputs are computed. The hook should have the following signature:: hook(module, grad_input, grad_output) -> tuple(Tensor) or None The :attr:`grad_input` and :attr:`grad_output` are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of :attr:`grad_input` in subsequent computations. :attr:`grad_input` will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in :attr:`grad_input` and :attr:`grad_output` will be ``None`` for all non-Tensor arguments. For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function. .. warning :: Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error. :param hook: The user-defined hook to be registered. :type hook: Callable :param prepend: If true, the provided ``hook`` will be fired before all existing ``backward`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``backward`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``backward`` hooks registered with :func:`register_module_full_backward_hook` will fire before all hooks registered by this method. :type prepend: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: _get_backward_hooks() Return the backward hooks for use in the call function. It returns two lists, one with the full backward hooks and one with the non-full backward hooks. .. py:method:: _get_backward_pre_hooks() .. py:method:: _maybe_warn_non_full_backward_hook(inputs, result, grad_fn) .. py:method:: register_forward_pre_hook(hook, *, prepend = False, with_kwargs = False) Register a forward pre-hook on the module. The hook will be called every time before :func:`forward` is invoked. If ``with_kwargs`` is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the ``forward``. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature:: hook(module, args) -> None or modified input If ``with_kwargs`` is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature:: hook(module, args, kwargs) -> None or a tuple of modified input and kwargs :param hook: The user defined hook to be registered. :type hook: Callable :param prepend: If true, the provided ``hook`` will be fired before all existing ``forward_pre`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``forward_pre`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``forward_pre`` hooks registered with :func:`register_module_forward_pre_hook` will fire before all hooks registered by this method. Default: ``False`` :type prepend: bool :param with_kwargs: If true, the ``hook`` will be passed the kwargs given to the forward function. Default: ``False`` :type with_kwargs: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: register_forward_hook(hook, *, prepend = False, with_kwargs = False, always_call = False) Register a forward hook on the module. The hook will be called every time after :func:`forward` has computed an output. If ``with_kwargs`` is ``False`` or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the ``forward``. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after :func:`forward` is called. The hook should have the following signature:: hook(module, args, output) -> None or modified output If ``with_kwargs`` is ``True``, the forward hook will be passed the ``kwargs`` given to the forward function and be expected to return the output possibly modified. The hook should have the following signature:: hook(module, args, kwargs, output) -> None or modified output :param hook: The user defined hook to be registered. :type hook: Callable :param prepend: If ``True``, the provided ``hook`` will be fired before all existing ``forward`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``forward`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``forward`` hooks registered with :func:`register_module_forward_hook` will fire before all hooks registered by this method. Default: ``False`` :type prepend: bool :param with_kwargs: If ``True``, the ``hook`` will be passed the kwargs given to the forward function. Default: ``False`` :type with_kwargs: bool :param always_call: If ``True`` the ``hook`` will be run regardless of whether an exception is raised while calling the Module. Default: ``False`` :type always_call: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: _slow_forward(*input, **kwargs) .. py:method:: _wrapped_call_impl(*args, **kwargs) .. py:method:: _call_impl(*args, **kwargs) .. py:attribute:: __call__ :type: Callable[Ellipsis, Any] .. py:method:: __getstate__() .. py:method:: __setstate__(state) .. py:method:: __getattr__(name) .. py:method:: __setattr__(name, value) .. py:method:: __delattr__(name) .. py:method:: _register_state_dict_hook(hook) Register a post-hook for the :meth:`~torch.nn.Module.state_dict` method. It should have the following signature:: hook(module, state_dict, prefix, local_metadata) -> None or state_dict The registered hooks can modify the ``state_dict`` inplace or return a new one. If a new ``state_dict`` is returned, it will only be respected if it is the root module that :meth:`~nn.Module.state_dict` is called from. .. py:method:: register_state_dict_post_hook(hook) Register a post-hook for the :meth:`~torch.nn.Module.state_dict` method. It should have the following signature:: hook(module, state_dict, prefix, local_metadata) -> None The registered hooks can modify the ``state_dict`` inplace. .. py:method:: register_state_dict_pre_hook(hook) Register a pre-hook for the :meth:`~torch.nn.Module.state_dict` method. It should have the following signature:: hook(module, prefix, keep_vars) -> None The registered hooks can be used to perform pre-processing before the ``state_dict`` call is made. .. py:method:: _save_to_state_dict(destination, prefix, keep_vars) Save module state to the `destination` dictionary. The `destination` dictionary will contain the state of the module, but not its descendants. This is called on every submodule in :meth:`~torch.nn.Module.state_dict`. In rare cases, subclasses can achieve class-specific behavior by overriding this method with custom logic. :param destination: a dict where state will be stored :type destination: dict :param prefix: the prefix for parameters and buffers used in this module :type prefix: str .. py:attribute:: T_destination .. py:method:: state_dict(*, destination: T_destination, prefix: str = ..., keep_vars: bool = ...) -> T_destination state_dict(*, prefix: str = ..., keep_vars: bool = ...) -> Dict[str, Any] Return a dictionary containing references to the whole state of the module. Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to ``None`` are not included. .. note:: The returned object is a shallow copy. It contains references to the module's parameters and buffers. .. warning:: Currently ``state_dict()`` also accepts positional arguments for ``destination``, ``prefix`` and ``keep_vars`` in order. However, this is being deprecated and keyword arguments will be enforced in future releases. .. warning:: Please avoid the use of argument ``destination`` as it is not designed for end-users. :param destination: If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an ``OrderedDict`` will be created and returned. Default: ``None``. :type destination: dict, optional :param prefix: a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ``''``. :type prefix: str, optional :param keep_vars: by default the :class:`~torch.Tensor` s returned in the state dict are detached from autograd. If it's set to ``True``, detaching will not be performed. Default: ``False``. :type keep_vars: bool, optional :Returns: **dict** -- a dictionary containing a whole state of the module Example:: >>> # xdoctest: +SKIP("undefined vars") >>> module.state_dict().keys() ['bias', 'weight'] .. py:method:: _register_load_state_dict_pre_hook(hook, with_module=False) See :meth:`~torch.nn.Module.register_load_state_dict_pre_hook` for details. A subtle difference is that if ``with_module`` is set to ``False``, then the hook will not take the ``module`` as the first argument whereas :meth:`~torch.nn.Module.register_load_state_dict_pre_hook` always takes the ``module`` as the first argument. :param hook: Callable hook that will be invoked before loading the state dict. :type hook: Callable :param with_module: Whether or not to pass the module instance to the hook as the first parameter. :type with_module: bool, optional .. py:method:: register_load_state_dict_pre_hook(hook) Register a pre-hook to be run before module's :meth:`~nn.Module.load_state_dict` is called. It should have the following signature:: hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None # noqa: B950 :param hook: Callable hook that will be invoked before loading the state dict. :type hook: Callable .. py:method:: register_load_state_dict_post_hook(hook) Register a post-hook to be run after module's :meth:`~nn.Module.load_state_dict` is called. It should have the following signature:: hook(module, incompatible_keys) -> None The ``module`` argument is the current module that this hook is registered on, and the ``incompatible_keys`` argument is a ``NamedTuple`` consisting of attributes ``missing_keys`` and ``unexpected_keys``. ``missing_keys`` is a ``list`` of ``str`` containing the missing keys and ``unexpected_keys`` is a ``list`` of ``str`` containing the unexpected keys. The given incompatible_keys can be modified inplace if needed. Note that the checks performed when calling :func:`load_state_dict` with ``strict=True`` are affected by modifications the hook makes to ``missing_keys`` or ``unexpected_keys``, as expected. Additions to either set of keys will result in an error being thrown when ``strict=True``, and clearing out both missing and unexpected keys will avoid an error. :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: _load_from_state_dict(state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) Copy parameters and buffers from :attr:`state_dict` into only this module, but not its descendants. This is called on every submodule in :meth:`~torch.nn.Module.load_state_dict`. Metadata saved for this module in input :attr:`state_dict` is provided as :attr:`local_metadata`. For state dicts without metadata, :attr:`local_metadata` is empty. Subclasses can achieve class-specific backward compatible loading using the version number at `local_metadata.get("version", None)`. Additionally, :attr:`local_metadata` can also contain the key `assign_to_params_buffers` that indicates whether keys should be assigned their corresponding tensor in the state_dict. .. note:: :attr:`state_dict` is not the same object as the input :attr:`state_dict` to :meth:`~torch.nn.Module.load_state_dict`. So it can be modified. :param state_dict: a dict containing parameters and persistent buffers. :type state_dict: dict :param prefix: the prefix for parameters and buffers used in this module :type prefix: str :param local_metadata: a dict containing the metadata for this module. See :type local_metadata: dict :param strict: whether to strictly enforce that the keys in :attr:`state_dict` with :attr:`prefix` match the names of parameters and buffers in this module :type strict: bool :param missing_keys: if ``strict=True``, add missing keys to this list :type missing_keys: list of str :param unexpected_keys: if ``strict=True``, add unexpected keys to this list :type unexpected_keys: list of str :param error_msgs: error messages should be added to this list, and will be reported together in :meth:`~torch.nn.Module.load_state_dict` :type error_msgs: list of str .. py:method:: load_state_dict(state_dict, strict = True, assign = False) Copy parameters and buffers from :attr:`state_dict` into this module and its descendants. If :attr:`strict` is ``True``, then the keys of :attr:`state_dict` must exactly match the keys returned by this module's :meth:`~torch.nn.Module.state_dict` function. .. warning:: If :attr:`assign` is ``True`` the optimizer must be created after the call to :attr:`load_state_dict` unless :func:`~torch.__future__.get_swap_module_params_on_conversion` is ``True``. :param state_dict: a dict containing parameters and persistent buffers. :type state_dict: dict :param strict: whether to strictly enforce that the keys in :attr:`state_dict` match the keys returned by this module's :meth:`~torch.nn.Module.state_dict` function. Default: ``True`` :type strict: bool, optional :param assign: When ``False``, the properties of the tensors in the current module are preserved while when ``True``, the properties of the Tensors in the state dict are preserved. The only exception is the ``requires_grad`` field of :class:`~torch.nn.Parameter`s for which the value from the module is preserved. Default: ``False`` :type assign: bool, optional :Returns: **``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields** -- * **missing_keys** is a list of str containing any keys that are expected by this module but missing from the provided ``state_dict``. * **unexpected_keys** is a list of str containing the keys that are not expected by this module but present in the provided ``state_dict``. .. note:: If a parameter or buffer is registered as ``None`` and its corresponding key exists in :attr:`state_dict`, :meth:`load_state_dict` will raise a ``RuntimeError``. .. py:method:: _named_members(get_members_fn, prefix='', recurse=True, remove_duplicate = True) Help yield various names + members of modules. .. py:method:: parameters(recurse = True) Return an iterator over module parameters. This is typically passed to an optimizer. :param recurse: if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module. :type recurse: bool :Yields: *Parameter* -- module parameter Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for param in model.parameters(): >>> print(type(param), param.size()) (20L,) (20L, 1L, 5L, 5L) .. py:method:: named_parameters(prefix = '', recurse = True, remove_duplicate = True) Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself. :param prefix: prefix to prepend to all parameter names. :type prefix: str :param recurse: if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module. :type recurse: bool :param remove_duplicate: whether to remove the duplicated parameters in the result. Defaults to True. :type remove_duplicate: bool, optional :Yields: *(str, Parameter)* -- Tuple containing the name and parameter Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for name, param in self.named_parameters(): >>> if name in ['bias']: >>> print(param.size()) .. py:method:: buffers(recurse = True) Return an iterator over module buffers. :param recurse: if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. :type recurse: bool :Yields: *torch.Tensor* -- module buffer Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for buf in model.buffers(): >>> print(type(buf), buf.size()) (20L,) (20L, 1L, 5L, 5L) .. py:method:: named_buffers(prefix = '', recurse = True, remove_duplicate = True) Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself. :param prefix: prefix to prepend to all buffer names. :type prefix: str :param recurse: if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True. :type recurse: bool, optional :param remove_duplicate: whether to remove the duplicated buffers in the result. Defaults to True. :type remove_duplicate: bool, optional :Yields: *(str, torch.Tensor)* -- Tuple containing the name and buffer Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for name, buf in self.named_buffers(): >>> if name in ['running_var']: >>> print(buf.size()) .. py:method:: children() Return an iterator over immediate children modules. :Yields: *Module* -- a child module .. py:method:: named_children() Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself. :Yields: *(str, Module)* -- Tuple containing a name and child module Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for name, module in model.named_children(): >>> if name in ['conv4', 'conv5']: >>> print(module) .. py:method:: modules() Return an iterator over all modules in the network. :Yields: *Module* -- a module in the network .. note:: Duplicate modules are returned only once. In the following example, ``l`` will be returned only once. Example:: >>> l = nn.Linear(2, 2) >>> net = nn.Sequential(l, l) >>> for idx, m in enumerate(net.modules()): ... print(idx, '->', m) 0 -> Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) ) 1 -> Linear(in_features=2, out_features=2, bias=True) .. py:method:: named_modules(memo = None, prefix = '', remove_duplicate = True) Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself. :param memo: a memo to store the set of modules already added to the result :param prefix: a prefix that will be added to the name of the module :param remove_duplicate: whether to remove the duplicated module instances in the result or not :Yields: *(str, Module)* -- Tuple of name and module .. note:: Duplicate modules are returned only once. In the following example, ``l`` will be returned only once. Example:: >>> l = nn.Linear(2, 2) >>> net = nn.Sequential(l, l) >>> for idx, m in enumerate(net.named_modules()): ... print(idx, '->', m) 0 -> ('', Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) )) 1 -> ('0', Linear(in_features=2, out_features=2, bias=True)) .. py:method:: train(mode = True) Set the module in training mode. This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`, etc. :param mode: whether to set training mode (``True``) or evaluation mode (``False``). Default: ``True``. :type mode: bool :Returns: **Module** -- self .. py:method:: eval() Set the module in evaluation mode. This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`, etc. This is equivalent with :meth:`self.train(False) `. See :ref:`locally-disable-grad-doc` for a comparison between `.eval()` and several similar mechanisms that may be confused with it. :Returns: **Module** -- self .. py:method:: requires_grad_(requires_grad = True) Change if autograd should record operations on parameters in this module. This method sets the parameters' :attr:`requires_grad` attributes in-place. This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training). See :ref:`locally-disable-grad-doc` for a comparison between `.requires_grad_()` and several similar mechanisms that may be confused with it. :param requires_grad: whether autograd should record operations on parameters in this module. Default: ``True``. :type requires_grad: bool :Returns: **Module** -- self .. py:method:: zero_grad(set_to_none = True) Reset gradients of all model parameters. See similar function under :class:`torch.optim.Optimizer` for more context. :param set_to_none: instead of setting to zero, set the grads to None. See :meth:`torch.optim.Optimizer.zero_grad` for details. :type set_to_none: bool .. py:method:: share_memory() See :meth:`torch.Tensor.share_memory_`. .. py:method:: _get_name() .. py:method:: extra_repr() Set the extra representation of the module. To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable. .. py:method:: __repr__() .. py:method:: __dir__() .. py:method:: _replicate_for_data_parallel() .. py:method:: compile(*args, **kwargs) Compile this Module's forward using :func:`torch.compile`. This Module's `__call__` method is compiled and all arguments are passed as-is to :func:`torch.compile`. See :func:`torch.compile` for details on the arguments for this function. .. py:class:: RelExtrModernBertConfig(pretrained_model_name_or_path, **kwargs) Bases: :py:obj:`medcat.components.addons.relation_extraction.config.RelExtrBaseConfig` Class for ModernBertConfig .. py:attribute:: name :value: 'modern-bert-config' .. py:attribute:: pretrained_model_name_or_path :value: 'answerdotai/ModernBERT-base' .. py:attribute:: hf_model_config :type: transformers.models.modernbert.ModernBertConfig .. py:method:: load(pretrained_model_name_or_path, relcat_config, **kwargs) :classmethod: .. py:method:: __init__(pretrained_model_name_or_path, **kwargs) .. py:attribute:: model_type :value: 'relcat' .. py:method:: to_dict() Serializes this instance to a Python dictionary. :Returns: **`Dict[str, Any]`** -- Dictionary of all the attributes that make up this configuration instance. .. py:method:: save(save_path) .. py:attribute:: base_config_key :type: str :value: '' .. py:attribute:: sub_configs :type: Dict[str, PretrainedConfig] .. py:attribute:: is_composition :type: bool :value: False .. py:attribute:: attribute_map :type: Dict[str, str] .. py:attribute:: base_model_tp_plan :type: Optional[Dict[str, Any]] :value: None .. py:attribute:: _auto_class :type: Optional[str] :value: None .. py:method:: __setattr__(key, value) .. py:method:: __getattribute__(key) .. py:attribute:: return_dict .. py:attribute:: output_hidden_states .. py:attribute:: output_attentions .. py:attribute:: torchscript .. py:attribute:: torch_dtype .. py:attribute:: use_bfloat16 .. py:attribute:: tf_legacy_loss .. py:attribute:: pruned_heads .. py:attribute:: tie_word_embeddings .. py:attribute:: chunk_size_feed_forward .. py:attribute:: is_encoder_decoder .. py:attribute:: is_decoder .. py:attribute:: cross_attention_hidden_size .. py:attribute:: add_cross_attention .. py:attribute:: tie_encoder_decoder .. py:attribute:: architectures .. py:attribute:: finetuning_task .. py:attribute:: id2label .. py:attribute:: label2id .. py:attribute:: tokenizer_class .. py:attribute:: prefix .. py:attribute:: bos_token_id .. py:attribute:: pad_token_id .. py:attribute:: eos_token_id .. py:attribute:: sep_token_id .. py:attribute:: decoder_start_token_id .. py:attribute:: task_specific_params .. py:attribute:: problem_type .. py:attribute:: _name_or_path :value: '' .. py:attribute:: _commit_hash .. py:attribute:: _attn_implementation_internal .. py:attribute:: _attn_implementation_autoset :value: False .. py:attribute:: transformers_version .. py:property:: name_or_path :type: str .. py:property:: use_return_dict :type: bool Whether or not return [`~utils.ModelOutput`] instead of tuples. :type: `bool` .. py:property:: num_labels :type: int The number of labels for classification models. :type: `int` .. py:property:: _attn_implementation .. py:method:: save_pretrained(save_directory, push_to_hub = False, **kwargs) Save a configuration object to the directory `save_directory`, so that it can be re-loaded using the [`~PretrainedConfig.from_pretrained`] class method. :param save_directory: Directory where the configuration JSON file will be saved (will be created if it does not exist). :type save_directory: `str` or `os.PathLike` :param push_to_hub: Whether or not to push your model to the Hugging Face model hub after saving it. You can specify the repository you want to push to with `repo_id` (will default to the name of `save_directory` in your namespace). :type push_to_hub: `bool`, *optional*, defaults to `False` :param kwargs: Additional key word arguments passed along to the [`~utils.PushToHubMixin.push_to_hub`] method. :type kwargs: `Dict[str, Any]`, *optional* .. py:method:: _set_token_in_kwargs(kwargs, token=None) :staticmethod: Temporary method to deal with `token` and `use_auth_token`. This method is to avoid apply the same changes in all model config classes that overwrite `from_pretrained`. Need to clean up `use_auth_token` in a follow PR. .. py:method:: from_pretrained(pretrained_model_name_or_path, cache_dir = None, force_download = False, local_files_only = False, token = None, revision = 'main', **kwargs) :classmethod: Instantiate a [`PretrainedConfig`] (or a derived class) from a pretrained model configuration. :param pretrained_model_name_or_path: This can be either: - a string, the *model id* of a pretrained model configuration hosted inside a model repo on huggingface.co. - a path to a *directory* containing a configuration file saved using the [`~PretrainedConfig.save_pretrained`] method, e.g., `./my_model_directory/`. - a path or url to a saved configuration JSON *file*, e.g., `./my_model_directory/configuration.json`. :type pretrained_model_name_or_path: `str` or `os.PathLike` :param cache_dir: Path to a directory in which a downloaded pretrained model configuration should be cached if the standard cache should not be used. :type cache_dir: `str` or `os.PathLike`, *optional* :param force_download: Whether or not to force to (re-)download the configuration files and override the cached versions if they exist. :type force_download: `bool`, *optional*, defaults to `False` :param resume_download: Deprecated and ignored. All downloads are now resumed by default when possible. Will be removed in v5 of Transformers. :param proxies: A dictionary of proxy servers to use by protocol or endpoint, e.g., `{'http': 'foo.bar:3128', 'http://hostname': 'foo.bar:4012'}.` The proxies are used on each request. :type proxies: `Dict[str, str]`, *optional* :param token: The token to use as HTTP bearer authorization for remote files. If `True`, or not specified, will use the token generated when running `huggingface-cli login` (stored in `~/.huggingface`). :type token: `str` or `bool`, *optional* :param revision: The specific model version to use. It can be a branch name, a tag name, or a commit id, since we use a git-based system for storing models and other artifacts on huggingface.co, so `revision` can be any identifier allowed by git. To test a pull request you made on the Hub, you can pass `revision="refs/pr/"`. :type revision: `str`, *optional*, defaults to `"main"` :param return_unused_kwargs: If `False`, then this function returns just the final configuration object. If `True`, then this functions returns a `Tuple(config, unused_kwargs)` where *unused_kwargs* is a dictionary consisting of the key/value pairs whose keys are not configuration attributes: i.e., the part of `kwargs` which has not been used to update `config` and is otherwise ignored. :type return_unused_kwargs: `bool`, *optional*, defaults to `False` :param subfolder: In case the relevant files are located inside a subfolder of the model repo on huggingface.co, you can specify the folder name here. :type subfolder: `str`, *optional*, defaults to `""` :param kwargs: The values in kwargs of any keys which are configuration attributes will be used to override the loaded values. Behavior concerning key/value pairs whose keys are *not* configuration attributes is controlled by the `return_unused_kwargs` keyword parameter. :type kwargs: `Dict[str, Any]`, *optional* :Returns: **[`PretrainedConfig`]** -- The configuration object instantiated from this pretrained model. Examples: ```python # We can't instantiate directly the base class *PretrainedConfig* so let's show the examples on a # derived class: BertConfig config = BertConfig.from_pretrained( "google-bert/bert-base-uncased" ) # Download configuration from huggingface.co and cache. config = BertConfig.from_pretrained( "./test/saved_model/" ) # E.g. config (or model) was saved using *save_pretrained('./test/saved_model/')* config = BertConfig.from_pretrained("./test/saved_model/my_configuration.json") config = BertConfig.from_pretrained("google-bert/bert-base-uncased", output_attentions=True, foo=False) assert config.output_attentions == True config, unused_kwargs = BertConfig.from_pretrained( "google-bert/bert-base-uncased", output_attentions=True, foo=False, return_unused_kwargs=True ) assert config.output_attentions == True assert unused_kwargs == {"foo": False} ``` .. py:method:: get_config_dict(pretrained_model_name_or_path, **kwargs) :classmethod: From a `pretrained_model_name_or_path`, resolve to a dictionary of parameters, to be used for instantiating a [`PretrainedConfig`] using `from_dict`. :param pretrained_model_name_or_path: The identifier of the pre-trained checkpoint from which we want the dictionary of parameters. :type pretrained_model_name_or_path: `str` or `os.PathLike` :Returns: **`Tuple[Dict, Dict]`** -- The dictionary(ies) that will be used to instantiate the configuration object. .. py:method:: _get_config_dict(pretrained_model_name_or_path, **kwargs) :classmethod: .. py:method:: from_dict(config_dict, **kwargs) :classmethod: Instantiates a [`PretrainedConfig`] from a Python dictionary of parameters. :param config_dict: Dictionary that will be used to instantiate the configuration object. Such a dictionary can be retrieved from a pretrained checkpoint by leveraging the [`~PretrainedConfig.get_config_dict`] method. :type config_dict: `Dict[str, Any]` :param kwargs: Additional parameters from which to initialize the configuration object. :type kwargs: `Dict[str, Any]` :Returns: **[`PretrainedConfig`]** -- The configuration object instantiated from those parameters. .. py:method:: from_json_file(json_file) :classmethod: Instantiates a [`PretrainedConfig`] from the path to a JSON file of parameters. :param json_file: Path to the JSON file containing the parameters. :type json_file: `str` or `os.PathLike` :Returns: **[`PretrainedConfig`]** -- The configuration object instantiated from that JSON file. .. py:method:: _dict_from_json_file(json_file) :classmethod: .. py:method:: __eq__(other) .. py:method:: __repr__() .. py:method:: __iter__() .. py:method:: to_diff_dict() Removes all attributes from config which correspond to the default config attributes for better readability and serializes to a Python dictionary. :Returns: **`Dict[str, Any]`** -- Dictionary of all the attributes that make up this configuration instance, .. py:method:: to_json_string(use_diff = True) Serializes this instance to a JSON string. :param use_diff: If set to `True`, only the difference between the config instance and the default `PretrainedConfig()` is serialized to JSON string. :type use_diff: `bool`, *optional*, defaults to `True` :Returns: **`str`** -- String containing all the attributes that make up this configuration instance in JSON format. .. py:method:: to_json_file(json_file_path, use_diff = True) Save this instance to a JSON file. :param json_file_path: Path to the JSON file in which this configuration instance's parameters will be saved. :type json_file_path: `str` or `os.PathLike` :param use_diff: If set to `True`, only the difference between the config instance and the default `PretrainedConfig()` is serialized to JSON file. :type use_diff: `bool`, *optional*, defaults to `True` .. py:method:: update(config_dict) Updates attributes of this class with attributes from `config_dict`. :param config_dict: Dictionary of attributes that should be updated for this class. :type config_dict: `Dict[str, Any]` .. py:method:: update_from_string(update_str) Updates attributes of this class with attributes from `update_str`. The expected format is ints, floats and strings as is, and for booleans use `true` or `false`. For example: "n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index" The keys to change have to already exist in the config object. :param update_str: String with attributes that should be updated for this class. :type update_str: `str` .. py:method:: dict_torch_dtype_to_str(d) Checks whether the passed dictionary and its nested dicts have a *torch_dtype* key and if it's not None, converts torch.dtype to a string of just the type. For example, `torch.float32` get converted into *"float32"* string, which can then be stored in the json format. .. py:method:: register_for_auto_class(auto_class='AutoConfig') :classmethod: Register this class with a given auto class. This should only be used for custom configurations as the ones in the library are already mapped with `AutoConfig`. This API is experimental and may have some slight breaking changes in the next releases. :param auto_class: The auto class to register this new configuration with. :type auto_class: `str` or `type`, *optional*, defaults to `"AutoConfig"` .. py:method:: _get_global_generation_defaults() :staticmethod: .. py:method:: _get_non_default_generation_parameters() Gets the non-default generation parameters on the PretrainedConfig instance .. py:method:: get_text_config(decoder=False) Returns the config that is meant to be used with text IO. On most models, it is the original config instance itself. On specific composite models, it is under a set of valid names. If `decoder` is set to `True`, then only search for decoder config names. .. py:method:: _create_repo(repo_id, private = None, token = None, repo_url = None, organization = None) Create the repo if needed, cleans up repo_id with deprecated kwargs `repo_url` and `organization`, retrieves the token. .. py:method:: _get_files_timestamps(working_dir) Returns the list of files with their last modification timestamp. .. py:method:: _upload_modified_files(working_dir, repo_id, files_timestamps, commit_message = None, token = None, create_pr = False, revision = None, commit_description = None) Uploads all modified files in `working_dir` to `repo_id`, based on `files_timestamps`. .. py:method:: push_to_hub(repo_id, use_temp_dir = None, commit_message = None, private = None, token = None, max_shard_size = '5GB', create_pr = False, safe_serialization = True, revision = None, commit_description = None, tags = None, **deprecated_kwargs) Upload the {object_files} to the 🤗 Model Hub. :param repo_id: The name of the repository you want to push your {object} to. It should contain your organization name when pushing to a given organization. :type repo_id: `str` :param use_temp_dir: Whether or not to use a temporary directory to store the files saved before they are pushed to the Hub. Will default to `True` if there is no directory named like `repo_id`, `False` otherwise. :type use_temp_dir: `bool`, *optional* :param commit_message: Message to commit while pushing. Will default to `"Upload {object}"`. :type commit_message: `str`, *optional* :param private: Whether to make the repo private. If `None` (default), the repo will be public unless the organization's default is private. This value is ignored if the repo already exists. :type private: `bool`, *optional* :param token: The token to use as HTTP bearer authorization for remote files. If `True`, will use the token generated when running `huggingface-cli login` (stored in `~/.huggingface`). Will default to `True` if `repo_url` is not specified. :type token: `bool` or `str`, *optional* :param max_shard_size: Only applicable for models. The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size lower than this size. If expressed as a string, needs to be digits followed by a unit (like `"5MB"`). We default it to `"5GB"` so that users can easily load models on free-tier Google Colab instances without any CPU OOM issues. :type max_shard_size: `int` or `str`, *optional*, defaults to `"5GB"` :param create_pr: Whether or not to create a PR with the uploaded files or directly commit. :type create_pr: `bool`, *optional*, defaults to `False` :param safe_serialization: Whether or not to convert the model weights in safetensors format for safer serialization. :type safe_serialization: `bool`, *optional*, defaults to `True` :param revision: Branch to push the uploaded files to. :type revision: `str`, *optional* :param commit_description: The description of the commit that will be created :type commit_description: `str`, *optional* :param tags: List of tags to push on the Hub. :type tags: `List[str]`, *optional* Examples: ```python from transformers import {object_class} {object} = {object_class}.from_pretrained("google-bert/bert-base-cased") # Push the {object} to your namespace with the name "my-finetuned-bert". {object}.push_to_hub("my-finetuned-bert") # Push the {object} to an organization with the name "my-finetuned-bert". {object}.push_to_hub("huggingface/my-finetuned-bert") ``` .. py:data:: logger .. py:class:: RelExtrModernBertModel(pretrained_model_name_or_path, relcat_config, model_config) Bases: :py:obj:`medcat.components.addons.relation_extraction.models.RelExtrBaseModel` ModernBertModel class for RelCAT .. py:attribute:: name :value: 'modernbertmodel_relcat' .. py:method:: __init__(pretrained_model_name_or_path, relcat_config, model_config) Class to hold the ModernBERT model + model_config :param pretrained_model_name_or_path: path to load the model from, this can be a HF model i.e: "bert-base-uncased", if left empty, it is normally assumed that a model is loaded from 'model.dat' using the RelCAT.load() method. So if you are initializing/training a model from scratch be sure to base it on some model. :type pretrained_model_name_or_path: str :param relcat_config: relcat config. :type relcat_config: ConfigRelCAT :param model_config: HF bert config for model. :type model_config: Union[RelExtrBaseConfig | RelExtrModernBertConfig] .. py:attribute:: relcat_config :type: medcat.config.config_rel_cat.ConfigRelCAT .. py:attribute:: model_config :type: medcat.components.addons.relation_extraction.modernbert.config.RelExtrModernBertConfig .. py:attribute:: pretrained_model_name_or_path :type: str .. py:attribute:: hf_model .. py:attribute:: drop_out .. py:method:: load_specific(pretrained_model_name_or_path, relcat_config, model_config, **kwargs) :classmethod: .. py:method:: _reinitialize_dense_and_frozen_layers(relcat_config) Reinitialize the dense layers of the model :param relcat_config: relcat config. :type relcat_config: ConfigRelCAT .. py:method:: forward(input_ids = None, attention_mask = None, token_type_ids = None, position_ids = None, head_mask = None, encoder_hidden_states = None, encoder_attention_mask = None, Q = None, e1_e2_start = None, pooled_output = None) Forward pass for the model :param input_ids: input token ids. Defaults to None. :type input_ids: torch.Tensor :param attention_mask: attention mask for the input ids. Defaults to None. :type attention_mask: torch.Tensor :param token_type_ids: token type ids for the input ids. Defaults to None. :type token_type_ids: torch.Tensor :param position_ids: The position IDs. Defaults to None. :type position_ids: Any :param head_mask: The head mask. Defaults to None. :type head_mask: Any :param encoder_hidden_states: Encoder hidden states. Defaults to None. :type encoder_hidden_states: Any :param encoder_attention_mask: Encoder attention mask. Defaults to None. :type encoder_attention_mask: Any :param Q: Q. Defaults to None. :type Q: Any :param e1_e2_start: Start and end indices for the entities in the input ids. Defaults to None. :type e1_e2_start: Any :param pooled_output: The pooled output. Defaults to None. :type pooled_output: Any :Returns: **Optional[tuple[torch.Tensor, torch.Tensor]]** -- Logits for the relation classification task. .. py:method:: output2logits(pooled_output, sequence_output, input_ids, e1_e2_start) :param pooled_output: embedding of the CLS token :type pooled_output: torch.Tensor :param sequence_output: hidden states/embeddings for each token in the input text :type sequence_output: torch.Tensor :param input_ids: input token ids. :type input_ids: torch.Tensor :param e1_e2_start: annotation tags token position :type e1_e2_start: torch.Tensor :Returns: **torch.Tensor** -- classification probabilities for each token. .. py:method:: load(pretrained_model_name_or_path, relcat_config, model_config) :classmethod: Load the model from the given path :param pretrained_model_name_or_path: path to load the model from. :type pretrained_model_name_or_path: str :param relcat_config: relcat config. :type relcat_config: ConfigRelCAT :param model_config: The model-specific config. :type model_config: RelExtrBaseConfig :returns: **RelExtrBaseModel** -- The loaded model. .. py:attribute:: fc1 :type: torch.nn.Linear .. py:attribute:: fc2 :type: torch.nn.Linear .. py:attribute:: fc3 :type: torch.nn.Linear .. py:attribute:: dump_patches :type: bool :value: False .. py:attribute:: _version :type: int :value: 1 This allows better BC support for :meth:`load_state_dict`. In :meth:`state_dict`, the version number will be saved as in the attribute `_metadata` of the returned state dict, and thus pickled. `_metadata` is a dictionary with keys that follow the naming convention of state dict. See ``_load_from_state_dict`` on how to use this information in loading. If new parameters/buffers are added/removed from a module, this number shall be bumped, and the module's `_load_from_state_dict` method can compare the version number and do appropriate changes if the state dict is from before the change. .. py:attribute:: training :type: bool .. py:attribute:: _parameters :type: Dict[str, Optional[torch.nn.parameter.Parameter]] .. py:attribute:: _buffers :type: Dict[str, Optional[torch.Tensor]] .. py:attribute:: _non_persistent_buffers_set :type: Set[str] .. py:attribute:: _backward_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _backward_hooks :type: Dict[int, Callable] .. py:attribute:: _is_full_backward_hook :type: Optional[bool] .. py:attribute:: _forward_hooks :type: Dict[int, Callable] .. py:attribute:: _forward_hooks_with_kwargs :type: Dict[int, bool] .. py:attribute:: _forward_hooks_always_called :type: Dict[int, bool] .. py:attribute:: _forward_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _forward_pre_hooks_with_kwargs :type: Dict[int, bool] .. py:attribute:: _state_dict_hooks :type: Dict[int, Callable] .. py:attribute:: _load_state_dict_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _state_dict_pre_hooks :type: Dict[int, Callable] .. py:attribute:: _load_state_dict_post_hooks :type: Dict[int, Callable] .. py:attribute:: _modules :type: Dict[str, Optional[Module]] .. py:attribute:: call_super_init :type: bool :value: False .. py:attribute:: _compiled_call_impl :type: Optional[Callable] :value: None .. py:method:: register_buffer(name, tensor, persistent = True) Add a buffer to the module. This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm's ``running_mean`` is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting :attr:`persistent` to ``False``. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's :attr:`state_dict`. Buffers can be accessed as attributes using given names. :param name: name of the buffer. The buffer can be accessed from this module using the given name :type name: str :param tensor: buffer to be registered. If ``None``, then operations that run on buffers, such as :attr:`cuda`, are ignored. If ``None``, the buffer is **not** included in the module's :attr:`state_dict`. :type tensor: Tensor or None :param persistent: whether the buffer is part of this module's :attr:`state_dict`. :type persistent: bool Example:: >>> # xdoctest: +SKIP("undefined vars") >>> self.register_buffer('running_mean', torch.zeros(num_features)) .. py:method:: register_parameter(name, param) Add a parameter to the module. The parameter can be accessed as an attribute using given name. :param name: name of the parameter. The parameter can be accessed from this module using the given name :type name: str :param param: parameter to be added to the module. If ``None``, then operations that run on parameters, such as :attr:`cuda`, are ignored. If ``None``, the parameter is **not** included in the module's :attr:`state_dict`. :type param: Parameter or None .. py:method:: add_module(name, module) Add a child module to the current module. The module can be accessed as an attribute using the given name. :param name: name of the child module. The child module can be accessed from this module using the given name :type name: str :param module: child module to be added to the module. :type module: Module .. py:method:: register_module(name, module) Alias for :func:`add_module`. .. py:method:: get_submodule(target) Return the submodule given by ``target`` if it exists, otherwise throw an error. For example, let's say you have an ``nn.Module`` ``A`` that looks like this: .. code-block:: text A( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) ) (The diagram shows an ``nn.Module`` ``A``. ``A`` has a nested submodule ``net_b``, which itself has two submodules ``net_c`` and ``linear``. ``net_c`` then has a submodule ``conv``.) To check whether or not we have the ``linear`` submodule, we would call ``get_submodule("net_b.linear")``. To check whether we have the ``conv`` submodule, we would call ``get_submodule("net_b.net_c.conv")``. The runtime of ``get_submodule`` is bounded by the degree of module nesting in ``target``. A query against ``named_modules`` achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, ``get_submodule`` should always be used. :param target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.) :Returns: **torch.nn.Module** -- The submodule referenced by ``target`` :raises AttributeError: If the target string references an invalid path or resolves to something that is not an ``nn.Module`` .. py:method:: set_submodule(target, module) Set the submodule given by ``target`` if it exists, otherwise throw an error. For example, let's say you have an ``nn.Module`` ``A`` that looks like this: .. code-block:: text A( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) ) (The diagram shows an ``nn.Module`` ``A``. ``A`` has a nested submodule ``net_b``, which itself has two submodules ``net_c`` and ``linear``. ``net_c`` then has a submodule ``conv``.) To overide the ``Conv2d`` with a new submodule ``Linear``, you would call ``set_submodule("net_b.net_c.conv", nn.Linear(33, 16))``. :param target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.) :param module: The module to set the submodule to. :raises ValueError: If the target string is empty :raises AttributeError: If the target string references an invalid path or resolves to something that is not an ``nn.Module`` .. py:method:: get_parameter(target) Return the parameter given by ``target`` if it exists, otherwise throw an error. See the docstring for ``get_submodule`` for a more detailed explanation of this method's functionality as well as how to correctly specify ``target``. :param target: The fully-qualified string name of the Parameter to look for. (See ``get_submodule`` for how to specify a fully-qualified string.) :Returns: **torch.nn.Parameter** -- The Parameter referenced by ``target`` :raises AttributeError: If the target string references an invalid path or resolves to something that is not an ``nn.Parameter`` .. py:method:: get_buffer(target) Return the buffer given by ``target`` if it exists, otherwise throw an error. See the docstring for ``get_submodule`` for a more detailed explanation of this method's functionality as well as how to correctly specify ``target``. :param target: The fully-qualified string name of the buffer to look for. (See ``get_submodule`` for how to specify a fully-qualified string.) :Returns: **torch.Tensor** -- The buffer referenced by ``target`` :raises AttributeError: If the target string references an invalid path or resolves to something that is not a buffer .. py:method:: get_extra_state() Return any extra state to include in the module's state_dict. Implement this and a corresponding :func:`set_extra_state` for your module if you need to store extra state. This function is called when building the module's `state_dict()`. Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes. :Returns: **object** -- Any extra state to store in the module's state_dict .. py:method:: set_extra_state(state) Set extra state contained in the loaded `state_dict`. This function is called from :func:`load_state_dict` to handle any extra state found within the `state_dict`. Implement this function and a corresponding :func:`get_extra_state` for your module if you need to store extra state within its `state_dict`. :param state: Extra state from the `state_dict` :type state: dict .. py:method:: _apply(fn, recurse=True) .. py:method:: apply(fn) Apply ``fn`` recursively to every submodule (as returned by ``.children()``) as well as self. Typical use includes initializing the parameters of a model (see also :ref:`nn-init-doc`). :param fn: function to be applied to each submodule :type fn: :class:`Module` -> None :Returns: **Module** -- self Example:: >>> @torch.no_grad() >>> def init_weights(m): >>> print(m) >>> if type(m) == nn.Linear: >>> m.weight.fill_(1.0) >>> print(m.weight) >>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2)) >>> net.apply(init_weights) Linear(in_features=2, out_features=2, bias=True) Parameter containing: tensor([[1., 1.], [1., 1.]], requires_grad=True) Linear(in_features=2, out_features=2, bias=True) Parameter containing: tensor([[1., 1.], [1., 1.]], requires_grad=True) Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) ) .. py:method:: cuda(device = None) Move all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: ipu(device = None) Move all model parameters and buffers to the IPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: xpu(device = None) Move all model parameters and buffers to the XPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: mtia(device = None) Move all model parameters and buffers to the MTIA. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on MTIA while being optimized. .. note:: This method modifies the module in-place. :param device: if specified, all parameters will be copied to that device :type device: int, optional :Returns: **Module** -- self .. py:method:: cpu() Move all model parameters and buffers to the CPU. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: type(dst_type) Casts all parameters and buffers to :attr:`dst_type`. .. note:: This method modifies the module in-place. :param dst_type: the desired type :type dst_type: type or string :Returns: **Module** -- self .. py:method:: float() Casts all floating point parameters and buffers to ``float`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: double() Casts all floating point parameters and buffers to ``double`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: half() Casts all floating point parameters and buffers to ``half`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: bfloat16() Casts all floating point parameters and buffers to ``bfloat16`` datatype. .. note:: This method modifies the module in-place. :Returns: **Module** -- self .. py:method:: to_empty(*, device, recurse = True) Move the parameters and buffers to the specified device without copying storage. :param device: The desired device of the parameters and buffers in this module. :type device: :class:`torch.device` :param recurse: Whether parameters and buffers of submodules should be recursively moved to the specified device. :type recurse: bool :Returns: **Module** -- self .. py:method:: to(device: Optional[torch._prims_common.DeviceLikeType] = ..., dtype: Optional[Module.to.dtype] = ..., non_blocking: bool = ...) -> typing_extensions.Self to(dtype: Module.to.dtype, non_blocking: bool = ...) -> typing_extensions.Self to(tensor: torch.Tensor, non_blocking: bool = ...) -> typing_extensions.Self Move and/or cast the parameters and buffers. This can be called as .. function:: to(device=None, dtype=None, non_blocking=False) :noindex: .. function:: to(dtype, non_blocking=False) :noindex: .. function:: to(tensor, non_blocking=False) :noindex: .. function:: to(memory_format=torch.channels_last) :noindex: Its signature is similar to :meth:`torch.Tensor.to`, but only accepts floating point or complex :attr:`dtype`\ s. In addition, this method will only cast the floating point or complex parameters and buffers to :attr:`dtype` (if given). The integral parameters and buffers will be moved :attr:`device`, if that is given, but with dtypes unchanged. When :attr:`non_blocking` is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices. See below for examples. .. note:: This method modifies the module in-place. :param device: the desired device of the parameters and buffers in this module :type device: :class:`torch.device` :param dtype: the desired floating point or complex dtype of the parameters and buffers in this module :type dtype: :class:`torch.dtype` :param tensor: Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module :type tensor: torch.Tensor :param memory_format: the desired memory format for 4D parameters and buffers in this module (keyword only argument) :type memory_format: :class:`torch.memory_format` :Returns: **Module** -- self Examples:: >>> # xdoctest: +IGNORE_WANT("non-deterministic") >>> linear = nn.Linear(2, 2) >>> linear.weight Parameter containing: tensor([[ 0.1913, -0.3420], [-0.5113, -0.2325]]) >>> linear.to(torch.double) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1913, -0.3420], [-0.5113, -0.2325]], dtype=torch.float64) >>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_CUDA1) >>> gpu1 = torch.device("cuda:1") >>> linear.to(gpu1, dtype=torch.half, non_blocking=True) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1914, -0.3420], [-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1') >>> cpu = torch.device("cpu") >>> linear.to(cpu) Linear(in_features=2, out_features=2, bias=True) >>> linear.weight Parameter containing: tensor([[ 0.1914, -0.3420], [-0.5112, -0.2324]], dtype=torch.float16) >>> linear = nn.Linear(2, 2, bias=None).to(torch.cdouble) >>> linear.weight Parameter containing: tensor([[ 0.3741+0.j, 0.2382+0.j], [ 0.5593+0.j, -0.4443+0.j]], dtype=torch.complex128) >>> linear(torch.ones(3, 2, dtype=torch.cdouble)) tensor([[0.6122+0.j, 0.1150+0.j], [0.6122+0.j, 0.1150+0.j], [0.6122+0.j, 0.1150+0.j]], dtype=torch.complex128) .. py:method:: register_full_backward_pre_hook(hook, prepend = False) Register a backward pre-hook on the module. The hook will be called every time the gradients for the module are computed. The hook should have the following signature:: hook(module, grad_output) -> tuple[Tensor] or None The :attr:`grad_output` is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of :attr:`grad_output` in subsequent computations. Entries in :attr:`grad_output` will be ``None`` for all non-Tensor arguments. For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function. .. warning :: Modifying inputs inplace is not allowed when using backward hooks and will raise an error. :param hook: The user-defined hook to be registered. :type hook: Callable :param prepend: If true, the provided ``hook`` will be fired before all existing ``backward_pre`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``backward_pre`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``backward_pre`` hooks registered with :func:`register_module_full_backward_pre_hook` will fire before all hooks registered by this method. :type prepend: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: register_backward_hook(hook) Register a backward hook on the module. This function is deprecated in favor of :meth:`~torch.nn.Module.register_full_backward_hook` and the behavior of this function will change in future versions. :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: register_full_backward_hook(hook, prepend = False) Register a backward hook on the module. The hook will be called every time the gradients with respect to a module are computed, i.e. the hook will execute if and only if the gradients with respect to module outputs are computed. The hook should have the following signature:: hook(module, grad_input, grad_output) -> tuple(Tensor) or None The :attr:`grad_input` and :attr:`grad_output` are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of :attr:`grad_input` in subsequent computations. :attr:`grad_input` will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in :attr:`grad_input` and :attr:`grad_output` will be ``None`` for all non-Tensor arguments. For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function. .. warning :: Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error. :param hook: The user-defined hook to be registered. :type hook: Callable :param prepend: If true, the provided ``hook`` will be fired before all existing ``backward`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``backward`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``backward`` hooks registered with :func:`register_module_full_backward_hook` will fire before all hooks registered by this method. :type prepend: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: _get_backward_hooks() Return the backward hooks for use in the call function. It returns two lists, one with the full backward hooks and one with the non-full backward hooks. .. py:method:: _get_backward_pre_hooks() .. py:method:: _maybe_warn_non_full_backward_hook(inputs, result, grad_fn) .. py:method:: register_forward_pre_hook(hook, *, prepend = False, with_kwargs = False) Register a forward pre-hook on the module. The hook will be called every time before :func:`forward` is invoked. If ``with_kwargs`` is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the ``forward``. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature:: hook(module, args) -> None or modified input If ``with_kwargs`` is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature:: hook(module, args, kwargs) -> None or a tuple of modified input and kwargs :param hook: The user defined hook to be registered. :type hook: Callable :param prepend: If true, the provided ``hook`` will be fired before all existing ``forward_pre`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``forward_pre`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``forward_pre`` hooks registered with :func:`register_module_forward_pre_hook` will fire before all hooks registered by this method. Default: ``False`` :type prepend: bool :param with_kwargs: If true, the ``hook`` will be passed the kwargs given to the forward function. Default: ``False`` :type with_kwargs: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: register_forward_hook(hook, *, prepend = False, with_kwargs = False, always_call = False) Register a forward hook on the module. The hook will be called every time after :func:`forward` has computed an output. If ``with_kwargs`` is ``False`` or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the ``forward``. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after :func:`forward` is called. The hook should have the following signature:: hook(module, args, output) -> None or modified output If ``with_kwargs`` is ``True``, the forward hook will be passed the ``kwargs`` given to the forward function and be expected to return the output possibly modified. The hook should have the following signature:: hook(module, args, kwargs, output) -> None or modified output :param hook: The user defined hook to be registered. :type hook: Callable :param prepend: If ``True``, the provided ``hook`` will be fired before all existing ``forward`` hooks on this :class:`torch.nn.modules.Module`. Otherwise, the provided ``hook`` will be fired after all existing ``forward`` hooks on this :class:`torch.nn.modules.Module`. Note that global ``forward`` hooks registered with :func:`register_module_forward_hook` will fire before all hooks registered by this method. Default: ``False`` :type prepend: bool :param with_kwargs: If ``True``, the ``hook`` will be passed the kwargs given to the forward function. Default: ``False`` :type with_kwargs: bool :param always_call: If ``True`` the ``hook`` will be run regardless of whether an exception is raised while calling the Module. Default: ``False`` :type always_call: bool :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: _slow_forward(*input, **kwargs) .. py:method:: _wrapped_call_impl(*args, **kwargs) .. py:method:: _call_impl(*args, **kwargs) .. py:attribute:: __call__ :type: Callable[Ellipsis, Any] .. py:method:: __getstate__() .. py:method:: __setstate__(state) .. py:method:: __getattr__(name) .. py:method:: __setattr__(name, value) .. py:method:: __delattr__(name) .. py:method:: _register_state_dict_hook(hook) Register a post-hook for the :meth:`~torch.nn.Module.state_dict` method. It should have the following signature:: hook(module, state_dict, prefix, local_metadata) -> None or state_dict The registered hooks can modify the ``state_dict`` inplace or return a new one. If a new ``state_dict`` is returned, it will only be respected if it is the root module that :meth:`~nn.Module.state_dict` is called from. .. py:method:: register_state_dict_post_hook(hook) Register a post-hook for the :meth:`~torch.nn.Module.state_dict` method. It should have the following signature:: hook(module, state_dict, prefix, local_metadata) -> None The registered hooks can modify the ``state_dict`` inplace. .. py:method:: register_state_dict_pre_hook(hook) Register a pre-hook for the :meth:`~torch.nn.Module.state_dict` method. It should have the following signature:: hook(module, prefix, keep_vars) -> None The registered hooks can be used to perform pre-processing before the ``state_dict`` call is made. .. py:method:: _save_to_state_dict(destination, prefix, keep_vars) Save module state to the `destination` dictionary. The `destination` dictionary will contain the state of the module, but not its descendants. This is called on every submodule in :meth:`~torch.nn.Module.state_dict`. In rare cases, subclasses can achieve class-specific behavior by overriding this method with custom logic. :param destination: a dict where state will be stored :type destination: dict :param prefix: the prefix for parameters and buffers used in this module :type prefix: str .. py:attribute:: T_destination .. py:method:: state_dict(*, destination: T_destination, prefix: str = ..., keep_vars: bool = ...) -> T_destination state_dict(*, prefix: str = ..., keep_vars: bool = ...) -> Dict[str, Any] Return a dictionary containing references to the whole state of the module. Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to ``None`` are not included. .. note:: The returned object is a shallow copy. It contains references to the module's parameters and buffers. .. warning:: Currently ``state_dict()`` also accepts positional arguments for ``destination``, ``prefix`` and ``keep_vars`` in order. However, this is being deprecated and keyword arguments will be enforced in future releases. .. warning:: Please avoid the use of argument ``destination`` as it is not designed for end-users. :param destination: If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an ``OrderedDict`` will be created and returned. Default: ``None``. :type destination: dict, optional :param prefix: a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ``''``. :type prefix: str, optional :param keep_vars: by default the :class:`~torch.Tensor` s returned in the state dict are detached from autograd. If it's set to ``True``, detaching will not be performed. Default: ``False``. :type keep_vars: bool, optional :Returns: **dict** -- a dictionary containing a whole state of the module Example:: >>> # xdoctest: +SKIP("undefined vars") >>> module.state_dict().keys() ['bias', 'weight'] .. py:method:: _register_load_state_dict_pre_hook(hook, with_module=False) See :meth:`~torch.nn.Module.register_load_state_dict_pre_hook` for details. A subtle difference is that if ``with_module`` is set to ``False``, then the hook will not take the ``module`` as the first argument whereas :meth:`~torch.nn.Module.register_load_state_dict_pre_hook` always takes the ``module`` as the first argument. :param hook: Callable hook that will be invoked before loading the state dict. :type hook: Callable :param with_module: Whether or not to pass the module instance to the hook as the first parameter. :type with_module: bool, optional .. py:method:: register_load_state_dict_pre_hook(hook) Register a pre-hook to be run before module's :meth:`~nn.Module.load_state_dict` is called. It should have the following signature:: hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None # noqa: B950 :param hook: Callable hook that will be invoked before loading the state dict. :type hook: Callable .. py:method:: register_load_state_dict_post_hook(hook) Register a post-hook to be run after module's :meth:`~nn.Module.load_state_dict` is called. It should have the following signature:: hook(module, incompatible_keys) -> None The ``module`` argument is the current module that this hook is registered on, and the ``incompatible_keys`` argument is a ``NamedTuple`` consisting of attributes ``missing_keys`` and ``unexpected_keys``. ``missing_keys`` is a ``list`` of ``str`` containing the missing keys and ``unexpected_keys`` is a ``list`` of ``str`` containing the unexpected keys. The given incompatible_keys can be modified inplace if needed. Note that the checks performed when calling :func:`load_state_dict` with ``strict=True`` are affected by modifications the hook makes to ``missing_keys`` or ``unexpected_keys``, as expected. Additions to either set of keys will result in an error being thrown when ``strict=True``, and clearing out both missing and unexpected keys will avoid an error. :Returns: **:class:`torch.utils.hooks.RemovableHandle`** -- a handle that can be used to remove the added hook by calling ``handle.remove()`` .. py:method:: _load_from_state_dict(state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) Copy parameters and buffers from :attr:`state_dict` into only this module, but not its descendants. This is called on every submodule in :meth:`~torch.nn.Module.load_state_dict`. Metadata saved for this module in input :attr:`state_dict` is provided as :attr:`local_metadata`. For state dicts without metadata, :attr:`local_metadata` is empty. Subclasses can achieve class-specific backward compatible loading using the version number at `local_metadata.get("version", None)`. Additionally, :attr:`local_metadata` can also contain the key `assign_to_params_buffers` that indicates whether keys should be assigned their corresponding tensor in the state_dict. .. note:: :attr:`state_dict` is not the same object as the input :attr:`state_dict` to :meth:`~torch.nn.Module.load_state_dict`. So it can be modified. :param state_dict: a dict containing parameters and persistent buffers. :type state_dict: dict :param prefix: the prefix for parameters and buffers used in this module :type prefix: str :param local_metadata: a dict containing the metadata for this module. See :type local_metadata: dict :param strict: whether to strictly enforce that the keys in :attr:`state_dict` with :attr:`prefix` match the names of parameters and buffers in this module :type strict: bool :param missing_keys: if ``strict=True``, add missing keys to this list :type missing_keys: list of str :param unexpected_keys: if ``strict=True``, add unexpected keys to this list :type unexpected_keys: list of str :param error_msgs: error messages should be added to this list, and will be reported together in :meth:`~torch.nn.Module.load_state_dict` :type error_msgs: list of str .. py:method:: load_state_dict(state_dict, strict = True, assign = False) Copy parameters and buffers from :attr:`state_dict` into this module and its descendants. If :attr:`strict` is ``True``, then the keys of :attr:`state_dict` must exactly match the keys returned by this module's :meth:`~torch.nn.Module.state_dict` function. .. warning:: If :attr:`assign` is ``True`` the optimizer must be created after the call to :attr:`load_state_dict` unless :func:`~torch.__future__.get_swap_module_params_on_conversion` is ``True``. :param state_dict: a dict containing parameters and persistent buffers. :type state_dict: dict :param strict: whether to strictly enforce that the keys in :attr:`state_dict` match the keys returned by this module's :meth:`~torch.nn.Module.state_dict` function. Default: ``True`` :type strict: bool, optional :param assign: When ``False``, the properties of the tensors in the current module are preserved while when ``True``, the properties of the Tensors in the state dict are preserved. The only exception is the ``requires_grad`` field of :class:`~torch.nn.Parameter`s for which the value from the module is preserved. Default: ``False`` :type assign: bool, optional :Returns: **``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields** -- * **missing_keys** is a list of str containing any keys that are expected by this module but missing from the provided ``state_dict``. * **unexpected_keys** is a list of str containing the keys that are not expected by this module but present in the provided ``state_dict``. .. note:: If a parameter or buffer is registered as ``None`` and its corresponding key exists in :attr:`state_dict`, :meth:`load_state_dict` will raise a ``RuntimeError``. .. py:method:: _named_members(get_members_fn, prefix='', recurse=True, remove_duplicate = True) Help yield various names + members of modules. .. py:method:: parameters(recurse = True) Return an iterator over module parameters. This is typically passed to an optimizer. :param recurse: if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module. :type recurse: bool :Yields: *Parameter* -- module parameter Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for param in model.parameters(): >>> print(type(param), param.size()) (20L,) (20L, 1L, 5L, 5L) .. py:method:: named_parameters(prefix = '', recurse = True, remove_duplicate = True) Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself. :param prefix: prefix to prepend to all parameter names. :type prefix: str :param recurse: if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module. :type recurse: bool :param remove_duplicate: whether to remove the duplicated parameters in the result. Defaults to True. :type remove_duplicate: bool, optional :Yields: *(str, Parameter)* -- Tuple containing the name and parameter Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for name, param in self.named_parameters(): >>> if name in ['bias']: >>> print(param.size()) .. py:method:: buffers(recurse = True) Return an iterator over module buffers. :param recurse: if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. :type recurse: bool :Yields: *torch.Tensor* -- module buffer Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for buf in model.buffers(): >>> print(type(buf), buf.size()) (20L,) (20L, 1L, 5L, 5L) .. py:method:: named_buffers(prefix = '', recurse = True, remove_duplicate = True) Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself. :param prefix: prefix to prepend to all buffer names. :type prefix: str :param recurse: if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True. :type recurse: bool, optional :param remove_duplicate: whether to remove the duplicated buffers in the result. Defaults to True. :type remove_duplicate: bool, optional :Yields: *(str, torch.Tensor)* -- Tuple containing the name and buffer Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for name, buf in self.named_buffers(): >>> if name in ['running_var']: >>> print(buf.size()) .. py:method:: children() Return an iterator over immediate children modules. :Yields: *Module* -- a child module .. py:method:: named_children() Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself. :Yields: *(str, Module)* -- Tuple containing a name and child module Example:: >>> # xdoctest: +SKIP("undefined vars") >>> for name, module in model.named_children(): >>> if name in ['conv4', 'conv5']: >>> print(module) .. py:method:: modules() Return an iterator over all modules in the network. :Yields: *Module* -- a module in the network .. note:: Duplicate modules are returned only once. In the following example, ``l`` will be returned only once. Example:: >>> l = nn.Linear(2, 2) >>> net = nn.Sequential(l, l) >>> for idx, m in enumerate(net.modules()): ... print(idx, '->', m) 0 -> Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) ) 1 -> Linear(in_features=2, out_features=2, bias=True) .. py:method:: named_modules(memo = None, prefix = '', remove_duplicate = True) Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself. :param memo: a memo to store the set of modules already added to the result :param prefix: a prefix that will be added to the name of the module :param remove_duplicate: whether to remove the duplicated module instances in the result or not :Yields: *(str, Module)* -- Tuple of name and module .. note:: Duplicate modules are returned only once. In the following example, ``l`` will be returned only once. Example:: >>> l = nn.Linear(2, 2) >>> net = nn.Sequential(l, l) >>> for idx, m in enumerate(net.named_modules()): ... print(idx, '->', m) 0 -> ('', Sequential( (0): Linear(in_features=2, out_features=2, bias=True) (1): Linear(in_features=2, out_features=2, bias=True) )) 1 -> ('0', Linear(in_features=2, out_features=2, bias=True)) .. py:method:: train(mode = True) Set the module in training mode. This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`, etc. :param mode: whether to set training mode (``True``) or evaluation mode (``False``). Default: ``True``. :type mode: bool :Returns: **Module** -- self .. py:method:: eval() Set the module in evaluation mode. This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`, etc. This is equivalent with :meth:`self.train(False) `. See :ref:`locally-disable-grad-doc` for a comparison between `.eval()` and several similar mechanisms that may be confused with it. :Returns: **Module** -- self .. py:method:: requires_grad_(requires_grad = True) Change if autograd should record operations on parameters in this module. This method sets the parameters' :attr:`requires_grad` attributes in-place. This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training). See :ref:`locally-disable-grad-doc` for a comparison between `.requires_grad_()` and several similar mechanisms that may be confused with it. :param requires_grad: whether autograd should record operations on parameters in this module. Default: ``True``. :type requires_grad: bool :Returns: **Module** -- self .. py:method:: zero_grad(set_to_none = True) Reset gradients of all model parameters. See similar function under :class:`torch.optim.Optimizer` for more context. :param set_to_none: instead of setting to zero, set the grads to None. See :meth:`torch.optim.Optimizer.zero_grad` for details. :type set_to_none: bool .. py:method:: share_memory() See :meth:`torch.Tensor.share_memory_`. .. py:method:: _get_name() .. py:method:: extra_repr() Set the extra representation of the module. To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable. .. py:method:: __repr__() .. py:method:: __dir__() .. py:method:: _replicate_for_data_parallel() .. py:method:: compile(*args, **kwargs) Compile this Module's forward using :func:`torch.compile`. This Module's `__call__` method is compiled and all arguments are passed as-is to :func:`torch.compile`. See :func:`torch.compile` for details on the arguments for this function.