hookeai.material_model_finder.model.material_discovery.MaterialModelFinder¶

class MaterialModelFinder(model_directory, model_name='material_model_finder', force_equilibrium_loss_type='pointwise', is_force_normalization=False, is_store_force_equilibrium_loss_hist=False, is_store_local_paths=False, local_paths_elements=None, is_compute_sets_reaction_hist=False, is_detect_autograd_anomaly=False, device_type='cpu')[source]¶

Bases: Module

Material model finder forward model.

_specimen_data¶

Specimen numerical data translated from experimental results.

Type:: SpecimenNumericalData

_specimen_material_state¶

FETorch specimen material state.

Type:: StructureMaterialState

_force_equilibrium_loss_type¶

Type of force equilibrium loss:

‘pointwise’Force equilibrium strictly based on pointwise: internal, external and reaction forces.
‘dirichlet_sets’Force equilibrium (1) based on pointwise: internal and external forces (non-Dirichlet degrees of freedom) and (2) based on pointwise internal, external and set-based reaction forces (Dirichlet constrained degrees of freedom).

Type:: str

_is_force_normalization¶

If True, then normalize forces prior to the computation of the force equilibrium loss.

Type:: bool, default=False

_data_scalers¶

Data scaler (item, TorchStandardScaler) for each feature data (key, str).

Type:: dict

_loss_scaling_factor¶

Loss scaling factor. If provided, then loss is pre-multiplied by loss scaling factor.

Type:: torch.Tensor(0d)

_loss_time_weights¶

Loss time weights stored as torch.Tensor(1d) of shape (n_time). If provided, then each discrete time loss contribution is pre-multiplied by corresponding weight. If None, time weights are set to 1.0.

Type:: torch.Tensor(1d), default=None

_is_store_force_equilibrium_loss_hist¶

If True, then store force equilibrium loss components history.

Type:: bool

_is_store_local_paths¶

If True, then store data set of specimen local (Gauss integration points) strain-stress paths in dedicated model subdirectory. Overwrites existing data set.

Type:: bool

_local_paths_elements¶

Elements for which local (Gauss integration points) strain-stress paths are stored as part of the specimen local data set. Elements are labeled from 1 to n_elem. If None, then all elements are stored. Only effective if is_store_local_paths=True.

Type:: list[int]

_is_compute_sets_reaction_hist¶

If True, then compute reaction forces history of Dirichlet boundary sets. Only available for ‘dirichlet_sets’ force equilibrium loss type.

Type:: bool

model_directory¶

Directory where model is stored.

Type:: str

model_name¶

Name of model.

Type:: str

_material_models_dir¶

Model subdirectory where material models are stored.

Type:: str

_internal_data_normalization_dir¶

Model subdirectory where the internal data normalization parameters are stored.

Type:: str

_temp_dir¶

Model subdirectory where temporary data is stored.

Type:: str

_device_type¶

Type of device on which torch.Tensor is allocated.

Type:: {‘cpu’, ‘cuda’}, default=’cpu’

_device¶

Device on which torch.Tensor is allocated.

Type:: torch.device

_set_model_subdirs(self)[source]¶: Set model subdirectories.

_set_material_models_dirs(self)[source]¶: Set material models directories.

check_force_equilibrium_loss_type(cls, force_equilibrium_loss_type)[source]¶: Check if force equilibrium loss type is available.

set_specimen_data(self, specimen_data, specimen_material_state,: force_minimum=None, force_maximum=None)

Set specimen data and material state.

get_detached_model_parameters(self)[source]¶: Get model parameters (material models) detached of gradients.

get_model_parameters_bounds(self)[source]¶: Get model parameters (material models) bounds.

enforce_parameters_bounds(self)[source]¶: Enforce bounds in model parameters (material models).

enforce_parameters_constraints(self)[source]¶: Enforce material model-dependent parameters constraints.

set_device(self, device_type)[source]¶: Set device on which torch.Tensor is allocated.

get_device(self)[source]¶: Get device on which torch.Tensor is allocated.

forward(self, sequential_mode='sequential_element')[source]¶: Forward propagation.

forward_sequential_time(self)[source]¶: Forward propagation (sequential time).

forward_sequential_element(self, is_store_local_paths=False)[source]¶: Forward propagation (sequential element).

compute_element_internal_forces_hist(self, strain_formulation, problem_type, element_type, element_material, element_state_old, nodes_coords_hist, nodes_disps_hist, nodes_inc_disps_hist, time_hist, is_recurrent_model)[source]¶: Compute history of finite element internal forces.

recurrent_material_state_update(self, strain_formulation, problem_type, constitutive_model, strain_hist, time_hist)[source]¶: Material state update for any given recurrent constitutive model.

force_equilibrium_loss(self, internal_forces_mesh, external_forces_mesh, reaction_forces_mesh, dirichlet_bool_mesh)[source]¶: Compute force equilibrium loss for given discrete time.

build_element_local_samples(self, strain_formulation, problem_type, element_type, time_hist, element_state_hist)[source]¶: Build element Gauss integration points local strain-stress paths.

compute_dirichlet_sets_reaction_hist(self, dirichlet_bc_mesh_hist, dirichlet_bool_mesh_hist)[source]¶: Compute Dirichlet boundary sets reaction forces history.

compute_dirichlet_sets_reaction(self, internal_forces_mesh, external_forces_mesh, dirichlet_bc_mesh)[source]¶: Compute reaction forces of Dirichlet boundary sets.

store_dirichlet_sets_reaction_hist(self, dirichlet_sets_reaction_hist, is_plot=True)[source]¶: Store reaction forces history of Dirichlet boundary sets.

build_tensor_from_comps(cls, n_dim, comps, comps_array, is_symmetric=False, device=None)[source]¶: Build strain/stress tensor from given components.

store_tensor_comps(cls, comps, tensor, device=None)[source]¶: Store strain/stress tensor components in array.

vforward_sequential_element(self)[source]¶: Forward propagation (sequential element).

vcompute_elements_internal_forces_hist(self, strain_formulation, problem_type, element_type, element_material, elements_coords_hist, elements_disps_hist, time_hist)[source]¶: Compute history of finite elements internal forces.

vcompute_element_internal_forces_hist(self, nodes_coords_hist, nodes_disps_hist, strain_formulation, problem_type, element_type, element_material, time_hist)[source]¶: Compute history of finite element internal forces.

vcompute_element_vol_grad_hist(self, nodes_coords_hist, nodes_disps_hist, strain_formulation, problem_type, element_type, time_hist)[source]¶: Compute history of finite element volumetric gradient operator.

vcompute_local_vol_grad_operator_hist(self, local_coords, weight, strain_formulation, problem_type, element_type, nodes_coords_hist, nodes_disps_hist, time_hist)[source]¶: Compute local integration point gradient contribution history.

vcompute_local_gradient(self, nodes_coords, local_coords, comp_order, element_type, is_symmetric=True)[source]¶: Compute discrete gradient operator at given local point of element.

vcompute_local_vol_sym_gradient(self, grad_operator_sym, n_dim)[source]¶: Compute discrete volumetric symmetric gradient operator.

vcompute_local_internal_forces_hist(self, local_coords, weight, strain_formulation, problem_type, element_type, nodes_coords_hist, nodes_disps_hist, time_hist, element_material, is_volumetric_bar=False, avg_vol_grad_operator_hist=None)[source]¶: Compute local integration point internal force contribution history.

vcompute_local_strain(self, nodes_coords, nodes_disps, local_coords, strain_formulation, n_dim, comp_order, element_type)[source]¶: Compute strain tensor at given local point of element.

vcompute_local_strain_vbar(self, nodes_coords, nodes_disps, avg_vol_grad_operator, local_coords, strain_formulation, n_dim, comp_order, element_type)[source]¶: Compute strain tensor at given local point of element.

vcompute_local_dev_sym_gradient(self, grad_operator_sym, n_dim)[source]¶: Compute discrete deviatoric symmetric gradient operator.

vrecurrent_material_state_update(self, strain_formulation, problem_type, constitutive_model, strain_hist, time_hist)[source]¶: Material state update for recurrent constitutive model.

vcompute_local_internal_forces(self, stress_vmf, grad_operator_sym, jacobian_det, weight)[source]¶: Compute local integration point internal forces contribution.

vbuild_internal_forces_mesh_hist(self, elements_internal_forces_hist, elements_mesh_indexes, n_node_mesh, n_dim)[source]¶: Build internal forces history of finite element mesh.

vassemble_internal_forces(self, elements_internal_forces, elements_mesh_indexes, n_node_mesh, n_dim)[source]¶: Assemble element internal forces into mesh counterpart.

vforce_equilibrium_hist_loss(self, internal_forces_mesh_hist, external_forces_mesh_hist, reaction_forces_mesh_hist, dirichlet_bc_mesh_hist)[source]¶: Compute force equilibrium history loss.

vforce_equilibrium_loss(self, internal_forces_mesh, external_forces_mesh, reaction_forces_mesh, dirichlet_bc_mesh)[source]¶: Compute force equilibrium loss.

force_equilibrium_loss_components_hist(self, internal_forces_mesh_hist, external_forces_mesh_hist, reaction_forces_mesh_hist, dirichlet_bc_mesh_hist)[source]¶: Compute force equilibrium loss components history (output purposes).

store_force_equilibrium_loss_components_hist(self, force_equilibrium_loss_components_hist, is_plot=True)[source]¶: Store force equilibrium loss components history.

build_elements_local_samples(self, strain_formulation, problem_type, time_hist, elements_state_hist)[source]¶: Build elements local strain-stress paths.

compute_dirichlet_sets_reaction_hist(self, internal_forces_mesh_hist, external_forces_mesh_hist, dirichlet_bc_mesh_hist)[source]¶: Compute reaction forces history of Dirichlet boundary sets.

compute_dirichlet_sets_reaction(self, internal_forces_mesh, external_forces_mesh, dirichlet_bc_mesh)[source]¶: Compute reaction forces of Dirichlet boundary sets.

store_dirichlet_sets_reaction_hist(self, dirichlet_sets_reaction_hist, is_export_csv=True, is_plot=True)[source]¶: Store reaction forces history of Dirichlet boundary sets.

vbuild_tensor_from_comps(cls, n_dim, comps, comps_array, device=None)[source]¶: Build strain/stress tensor from given components.

vstore_tensor_comps(cls, comps, tensor, device=None)[source]¶: Store strain/stress tensor components in array.

features_out_extractor(cls, model_output)[source]¶: Extract output features from generic model output.

_init_data_scalers(self)[source]¶: Initialize model data scalers.

set_fitted_force_data_scalers(self, force_minimum, force_maximum)[source]¶: Set fitted forces data scalers.

set_material_models_fitted_data_scalers(self, models_scaling_type, models_scaling_parameters)[source]¶

check_model_in_normalized(cls, model)[source]¶: Check if generic model expects normalized input features.

check_model_out_normalized(cls, model)[source]¶: Check if generic model expects normalized output features.

Constructor.

Parameters:

model_directory (str) – Directory where model is stored.
model_name (str, default='material_model_finder') – Name of model.
force_equilibrium_loss_type (str, default='pointwise') –
Type of force equilibrium loss:

’pointwise’Force equilibrium strictly based on pointwise
internal, external and reaction forces.

’dirichlet_sets’Force equilibrium (1) based on pointwise
internal and external forces (non-Dirichlet degrees of freedom) and (2) based on pointwise internal, external and set-based reaction forces (Dirichlet constrained degrees of freedom).
is_force_normalization (bool, default=False) – If True, then normalize forces prior to the computation of the force equilibrium loss.
is_store_force_equilibrium_loss_hist (bool, default=False) – If True, then store force equilibrium loss components history.
is_store_local_paths (bool, default=False) – If True, then store data set of specimen local (Gauss integration points) strain-stress paths in dedicated model subdirectory. Overwrites existing data set.
local_paths_elements (list[int], default=None) – Elements for which local (Gauss integration points) strain-stress paths are stored as part of the specimen local data set. Elements are labeled from 1 to n_elem. If None, then all elements are stored. Only effective if is_store_local_paths=True.
is_compute_sets_reaction_hist (bool, default=False) – If True, then compute reaction forces history of Dirichlet boundary sets. Only available for ‘dirichlet_sets’ force equilibrium loss type.
is_detect_autograd_anomaly (bool, default=False) – If True, then set context-manager that enables anomaly detection for the autograd engine. Should only be enabled for debugging purposes as it degrades performance.
device_type ({'cpu', 'cuda'}, default='cpu') – Type of device on which torch.Tensor is allocated.

List of Public Methods

`add_module`	Adds a child module to the current module.
`apply`	Applies `fn` recursively to every submodule (as returned by `.children()`) as well as self.
`bfloat16`	Casts all floating point parameters and buffers to `bfloat16` datatype.
`buffers`	Returns an iterator over module buffers.
`build_element_local_samples`	Build element Gauss integration points local strain-stress paths.
`build_elements_local_samples`	Build elements local strain-stress paths.
`build_tensor_from_comps`	Build strain/stress tensor from given components.
`check_force_equilibrium_loss_type`	Check if force equilibrium loss type is available.
`check_model_in_normalized`	Check if generic model expects normalized input features.
`check_model_out_normalized`	Check if generic model expects normalized output features.
`children`	Returns an iterator over immediate children modules.
`compile`	Compile this Module's forward using `torch.compile()`.
`compute_dirichlet_sets_reaction`	Compute reaction forces of Dirichlet boundary sets.
`compute_dirichlet_sets_reaction_hist`	Compute reaction forces history of Dirichlet boundary sets.
`compute_element_internal_forces_hist`	Compute history of finite element internal forces.
`cpu`	Moves all model parameters and buffers to the CPU.
`cuda`	Moves all model parameters and buffers to the GPU.
`double`	Casts all floating point parameters and buffers to `double` datatype.
`enforce_parameters_bounds`	Enforce bounds in model parameters (material models).
`enforce_parameters_constraints`	Enforce material model-dependent parameters constraints.
`eval`	Sets the module in evaluation mode.
`extra_repr`	Set the extra representation of the module
`features_out_extractor`	Extract output features from generic model output.
`float`	Casts all floating point parameters and buffers to `float` datatype.
`force_equilibrium_loss`	Compute force equilibrium loss.
`force_equilibrium_loss_components_hist`	Compute force equilibrium loss components history (output purposes).
`forward`	Forward propagation.
`forward_sequential_element`	Forward propagation (sequential element).
`forward_sequential_time`	Forward propagation (sequential time).
`get_buffer`	Returns the buffer given by `target` if it exists, otherwise throws an error.
`get_detached_model_parameters`	Get model parameters (material models) detached of gradients.
`get_device`	Get device on which torch.Tensor is allocated.
`get_extra_state`	Returns any extra state to include in the module's state_dict.
`get_material_models`	Get material models.
`get_model_parameters_bounds`	Get model parameters (material models) bounds.
`get_parameter`	Returns the parameter given by `target` if it exists, otherwise throws an error.
`get_submodule`	Returns the submodule given by `target` if it exists, otherwise throws an error.
`half`	Casts all floating point parameters and buffers to `half` datatype.
`ipu`	Moves all model parameters and buffers to the IPU.
`load_state_dict`	Copies parameters and buffers from `state_dict` into this module and its descendants.
`modules`	Returns an iterator over all modules in the network.
`named_buffers`	Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
`named_children`	Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
`named_modules`	Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
`named_parameters`	Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
`parameters`	Returns an iterator over module parameters.
`recurrent_material_state_update`	Material state update for any given recurrent constitutive model.
`register_backward_hook`	Registers a backward hook on the module.
`register_buffer`	Adds a buffer to the module.
`register_forward_hook`	Registers a forward hook on the module.
`register_forward_pre_hook`	Registers a forward pre-hook on the module.
`register_full_backward_hook`	Registers a backward hook on the module.
`register_full_backward_pre_hook`	Registers a backward pre-hook on the module.
`register_load_state_dict_post_hook`	Registers a post hook to be run after module's `load_state_dict` is called.
`register_module`	Alias for `add_module()`.
`register_parameter`	Adds a parameter to the module.
`register_state_dict_pre_hook`	These hooks will be called with arguments: `self`, `prefix`, and `keep_vars` before calling `state_dict` on `self`.
`requires_grad_`	Change if autograd should record operations on parameters in this module.
`set_device`	Set device on which torch.Tensor is allocated.
`set_extra_state`	This function is called from `load_state_dict()` to handle any extra state found within the state_dict.
`set_fitted_force_data_scalers`	Set fitted forces data scalers.
`set_material_models_fitted_data_scalers`	Set material constitutive models fitted data scalers.
`set_specimen_data`	Set specimen data and material state.
`share_memory`	See `torch.Tensor.share_memory_()`
`state_dict`	Returns a dictionary containing references to the whole state of the module.
`store_dirichlet_sets_reaction_hist`	Store reaction forces history of Dirichlet boundary sets.
`store_force_equilibrium_loss_components_hist`	Store force equilibrium loss components history.
`store_tensor_comps`	Store strain/stress tensor components in array.
`to`	Moves and/or casts the parameters and buffers.
`to_empty`	Moves the parameters and buffers to the specified device without copying storage.
`train`	Sets the module in training mode.
`type`	Casts all parameters and buffers to `dst_type`.
`vassemble_internal_forces`	Assemble element internal forces into mesh counterpart.
`vbuild_internal_forces_mesh_hist`	Build internal forces history of finite element mesh.
`vbuild_tensor_from_comps`	Build strain/stress tensor from given components.
`vcompute_element_internal_forces_hist`	Compute history of finite element internal forces.
`vcompute_element_vol_grad_hist`	Compute history of finite element volumetric gradient operator.
`vcompute_elements_internal_forces_hist`	Compute history of finite elements internal forces.
`vcompute_local_dev_sym_gradient`	Compute discrete deviatoric symmetric gradient operator.
`vcompute_local_gradient`	Compute discrete gradient operator at given local point of element.
`vcompute_local_internal_forces`	Compute local integration point internal forces contribution.
`vcompute_local_internal_forces_hist`	Compute local integration point internal force contribution history.
`vcompute_local_strain`	Compute strain tensor at given local point of element.
`vcompute_local_strain_vbar`	Compute strain tensor at given local point of element.
`vcompute_local_vol_grad_operator_hist`	Compute local integration point gradient contribution history.
`vcompute_local_vol_sym_gradient`	Compute discrete volumetric symmetric gradient operator.
`vforce_equilibrium_hist_loss`	Compute force equilibrium history loss.
`vforce_equilibrium_loss`	Compute force equilibrium loss.
`vforward_sequential_element`	Forward propagation (sequential element).
`vrecurrent_material_state_update`	Material state update for recurrent constitutive model.
`vstore_tensor_comps`	Store strain/stress tensor components in array.
`xpu`	Moves all model parameters and buffers to the XPU.
`zero_grad`	Resets gradients of all model parameters.

Attributes

`T_destination`
`call_super_init`
`dump_patches`
`training`

Methods

__call__(*args, **kwargs)¶: Call self as a function.

__init__(model_directory, model_name='material_model_finder', force_equilibrium_loss_type='pointwise', is_force_normalization=False, is_store_force_equilibrium_loss_hist=False, is_store_local_paths=False, local_paths_elements=None, is_compute_sets_reaction_hist=False, is_detect_autograd_anomaly=False, device_type='cpu')[source]¶

Constructor.

Parameters:

model_directory (str) – Directory where model is stored.
model_name (str, default='material_model_finder') – Name of model.
force_equilibrium_loss_type (str, default='pointwise') –
Type of force equilibrium loss:

’pointwise’Force equilibrium strictly based on pointwise
internal, external and reaction forces.

’dirichlet_sets’Force equilibrium (1) based on pointwise
internal and external forces (non-Dirichlet degrees of freedom) and (2) based on pointwise internal, external and set-based reaction forces (Dirichlet constrained degrees of freedom).
is_force_normalization (bool, default=False) – If True, then normalize forces prior to the computation of the force equilibrium loss.
is_store_force_equilibrium_loss_hist (bool, default=False) – If True, then store force equilibrium loss components history.
is_store_local_paths (bool, default=False) – If True, then store data set of specimen local (Gauss integration points) strain-stress paths in dedicated model subdirectory. Overwrites existing data set.
local_paths_elements (list[int], default=None) – Elements for which local (Gauss integration points) strain-stress paths are stored as part of the specimen local data set. Elements are labeled from 1 to n_elem. If None, then all elements are stored. Only effective if is_store_local_paths=True.
is_compute_sets_reaction_hist (bool, default=False) – If True, then compute reaction forces history of Dirichlet boundary sets. Only available for ‘dirichlet_sets’ force equilibrium loss type.
is_detect_autograd_anomaly (bool, default=False) – If True, then set context-manager that enables anomaly detection for the autograd engine. Should only be enabled for debugging purposes as it degrades performance.
device_type ({'cpu', 'cuda'}, default='cpu') – Type of device on which torch.Tensor is allocated.

_get_backward_hooks()¶: Returns 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.

_init_data_scalers()[source]¶: Initialize model data scalers.

_load_from_state_dict(state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs)¶

Copies parameters and buffers from state_dict into only this module, but not its descendants. This is called on every submodule in load_state_dict(). Metadata saved for this module in input state_dict is provided as local_metadata. For state dicts without metadata, local_metadata is empty. Subclasses can achieve class-specific backward compatible loading using the version number at local_metadata.get(“version”, None). Additionally, 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

state_dict is not the same object as the input state_dict to load_state_dict(). So it can be modified.

Parameters:

state_dict (dict) – a dict containing parameters and persistent buffers.
prefix (str) – the prefix for parameters and buffers used in this module
local_metadata (dict) – a dict containing the metadata for this module. See
strict (bool) – whether to strictly enforce that the keys in state_dict with prefix match the names of parameters and buffers in this module
missing_keys (list of str) – if strict=True, add missing keys to this list
unexpected_keys (list of str) – if strict=True, add unexpected keys to this list
error_msgs (list of str) – error messages should be added to this list, and will be reported together in load_state_dict()

_named_members(get_members_fn, prefix='', recurse=True, remove_duplicate=True)¶: Helper method for yielding various names + members of modules.

_register_load_state_dict_pre_hook(hook, with_module=False)¶

These hooks will be called with arguments: state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs, before loading state_dict into self. These arguments are exactly the same as those of _load_from_state_dict.

If with_module is True, then the first argument to the hook is an instance of the module.

Parameters:

hook (Callable) – Callable hook that will be invoked before loading the state dict.
with_module (bool, optional) – Whether or not to pass the module instance to the hook as the first parameter.

_register_state_dict_hook(hook)¶: These hooks will be called with arguments: self, state_dict, prefix, local_metadata, after the state_dict of self is set. Note that only parameters and buffers of self or its children are guaranteed to exist in state_dict. The hooks may modify state_dict inplace or return a new one.

_save_to_state_dict(destination, prefix, keep_vars)¶

Saves module state to destination dictionary, containing a state of the module, but not its descendants. This is called on every submodule in state_dict().

In rare cases, subclasses can achieve class-specific behavior by overriding this method with custom logic.

Parameters:

destination (dict) – a dict where state will be stored
prefix (str) – the prefix for parameters and buffers used in this module

_set_material_models_dirs()[source]¶: Set material models directories.

_set_model_subdirs()[source]¶: Set model subdirectories.

_version: int = 1¶

This allows better BC support for load_state_dict(). In 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.

add_module(name, module)¶

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters:

name (str) – name of the child module. The child module can be accessed from this module using the given name
module (Module) – child module to be added to the module.

Return type:

None

apply(fn)¶

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also nn-init-doc).

Parameters:: fn (Module -> None) – function to be applied to each submodule
Returns:: self
Return type:: Module

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)
)

bfloat16()¶

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns:: self
Return type:: Module

buffers(recurse=True)¶

Returns an iterator over module buffers.

Parameters:: recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.
Yields:: torch.Tensor – module buffer
Return type:: Iterator[Tensor]

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

build_element_local_samples(strain_formulation, problem_type, element_type, time_hist, element_state_hist)[source]¶

Build element Gauss integration points local strain-stress paths.

Parameters:

strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
element_type (Element) – FETorch finite element.
time_hist (torch.Tensor(1d)) – Discrete time history.
element_state_hist (list[dict]) – Material constitutive model state variables history (item, dict) for each Gauss integration point (key, str[int]).

Returns:

element_local_samples – Element local strain-stress paths, each corresponding to a given element Gauss integration point. Each path is stored as a dictionary where each feature (key, str) data is a torch.Tensor(2d) of shape (sequence_length, n_features).

Return type:

list[dict]

build_elements_local_samples(strain_formulation, problem_type, time_hist, elements_state_hist)[source]¶

Build elements local strain-stress paths.

Parameters:

strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
time_hist (torch.Tensor(1d)) – Discrete time history.
elements_state_hist (torch.Tensor(4d)) – Gauss integration points strain and stress path history of finite elements stored as torch.Tensor(4d) of shape (n_elem, n_gauss, n_time, n_strain_comps + n_stress_comps).

Returns:

elements_local_samples – Elements local strain-stress paths, each corresponding to a given element Gauss integration point. Each path is stored as a dictionary where each feature (key, str) data is a torch.Tensor(2d) of shape (sequence_length, n_features).

Return type:

list[dict]

classmethod build_tensor_from_comps(n_dim, comps, comps_array, is_symmetric=False, device=None)[source]¶

Build strain/stress tensor from given components.

Parameters:

n_dim (int) – Problem number of spatial dimensions.
comps (tuple[str]) – Strain/Stress components order.
comps_array (torch.Tensor(1d)) – Strain/Stress components array.
is_symmetric (bool, default=False) – If True, then assembles off-diagonal strain components from symmetric component.
device (torch.device, default=None) – Device on which torch.Tensor is allocated.

Returns:

tensor – Strain/Stress tensor.

Return type:

torch.Tensor(2d)

classmethod check_force_equilibrium_loss_type(force_equilibrium_loss_type)[source]¶

Check if force equilibrium loss type is available.

Parameters:: force_equilibrium_loss_type (str) – Type of force equilibrium loss.
Returns:: force_equilibrium_loss_type – Type of force equilibrium loss.
Return type:: str

classmethod check_model_in_normalized(model)[source]¶

Check if generic model expects normalized input features.

A model expects normalized input features if it has an attribute ‘is_model_in_normalized’ set to True.

Parameters:: model (torch.nn.Module) – Model.
Returns:: is_model_in_normalized – If True, then model expects normalized input features (normalized input data has been seen during model training).
Return type:: bool

classmethod check_model_out_normalized(model)[source]¶

Check if generic model expects normalized output features.

A model expects normalized output features if it has an attribute ‘is_model_out_normalized’ set to True.

Parameters:: model (torch.nn.Module) – Model.
Returns:: is_model_out_normalized – If True, then model expects normalized output features (normalized output data has been seen during model training).
Return type:: bool

children()¶

Returns an iterator over immediate children modules.

Yields:: Module – a child module
Return type:: Iterator[Module]

compile(*args, **kwargs)¶

Compile this Module’s forward using torch.compile().

This Module’s __call__ method is compiled and all arguments are passed as-is to torch.compile().

See torch.compile() for details on the arguments for this function.

compute_dirichlet_sets_reaction(internal_forces_mesh, external_forces_mesh, dirichlet_bc_mesh)[source]¶

Compute reaction forces of Dirichlet boundary sets.

Parameters:

internal_forces_mesh (torch.Tensor(2d)) – Internal forces of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).
external_forces_mesh (torch.Tensor(2d)) – External forces of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim). (n_node_mesh, n_dim).
dirichlet_bc_mesh (torch.Tensor(2d)) – Dirichlet boundary constraints of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim). Encodes if each degree of freedom is free (assigned 0) or constrained (greater than 0) under Dirichlet boundary conditions. The encoding depends on the selected force equilibrium loss type.

Returns:

dirichlet_sets_reaction – Reaction forces of Dirichlet boundary sets stored as torch.Tensor(2d) of shape (n_sets, 1). Sets are sorted according with their encoding labels and are associated with a single spatial dimension.

Return type:

torch.Tensor(2d)

compute_dirichlet_sets_reaction_hist(internal_forces_mesh_hist, external_forces_mesh_hist, dirichlet_bc_mesh_hist)[source]¶

Compute reaction forces history of Dirichlet boundary sets.

At a given time step, the reaction force of each Dirichlet boundary set is computed to strictly satisfy the total force equilibrium, assuming that the internal and external forces are known.

Parameters:

internal_forces_mesh_hist (torch.Tensor(3d)) – Internal forces history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
external_forces_mesh_hist (torch.Tensor(3d)) – External forces history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
dirichlet_bc_mesh_hist (torch.Tensor(3d)) – Dirichlet boundary constraints history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time). Encodes if each degree of freedom is free (assigned 0) or constrained (greater than 0) under Dirichlet boundary conditions. The encoding depends on the selected force equilibrium loss type.

Returns:

dirichlet_sets_reaction_hist – Reaction forces history of Dirichlet boundary sets stored as torch.Tensor(3d) of shape (n_sets, 1, n_time). Sets are sorted according with their encoding labels and are associated with a single spatial dimension.

Return type:

torch.Tensor(3d)

compute_element_internal_forces_hist(strain_formulation, problem_type, element_type, element_material, element_state_old, nodes_coords_hist, nodes_disps_hist, nodes_inc_disps_hist, time_hist, is_recurrent_model)[source]¶

Compute history of finite element internal forces.

Parameters:

strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
element_type (Element) – FETorch finite element.
element_material (ConstitutiveModel) – FETorch material constitutive model.
element_state_old (dict) – Last converged material constitutive model state variables (item, dict) for each Gauss integration point (key, str[int]).
nodes_coords_hist (torch.Tensor(3d)) – Coordinates history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
nodes_disps_hist (torch.Tensor(3d)) – Displacements history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
nodes_inc_disps_hist (torch.Tensor(3d)) – Incremental displacements history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
time_hist (torch.Tensor(1d)) – Discrete time history.
is_recurrent_model (bool) – True if the material constitutive model has a recurrent structure (processes full deformation path when called), False otherwise.

Returns:

element_internal_forces_hist (torch.Tensor(2d)) – Element internal forces history stored as torch.Tensor(2d) of shape (n_node*n_dof_node, n_time).
element_state_hist (list[dict]) – Material constitutive model state variables history (item, dict) for each Gauss integration point (key, str[int]).

cpu()¶

Moves all model parameters and buffers to the CPU.

Note

This method modifies the module in-place.

Returns:: self
Return type:: Module

cuda(device=None)¶

Moves 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.

Parameters:: device (int, optional) – if specified, all parameters will be copied to that device
Returns:: self
Return type:: Module

double()¶

Casts all floating point parameters and buffers to double datatype.

Note

This method modifies the module in-place.

Returns:: self
Return type:: Module

enforce_parameters_bounds()[source]¶

Enforce bounds in model parameters (material models).

Only enforces bounds in parameters from material models with explicit learnable parameters.

Bounds are enforced by means of in-place parameters updates.

enforce_parameters_constraints()[source]¶

Enforce material model-dependent parameters constraints.

Only enforces constraints in parameters from material models with explicit learnable parameters.

Constraints are enforced by means of in-place parameters updates.

eval()¶

Sets 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. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns:: self
Return type:: Module

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.

Return type:: str

classmethod features_out_extractor(model_output)[source]¶

Extract output features from generic model output.

Parameters:: model_output ({torch.Tensor, tuple}) – Model output.
Returns:: features_out – Tensor of output features stored as torch.Tensor(2d) of shape (sequence_length, n_features_out) for unbatched input or torch.Tensor(3d) of shape (sequence_length, batch_size, n_features_out) for batched input.
Return type:: torch.Tensor

float()¶

Casts all floating point parameters and buffers to float datatype.

Note

This method modifies the module in-place.

Returns:: self
Return type:: Module

force_equilibrium_loss(internal_forces_mesh, external_forces_mesh, reaction_forces_mesh, dirichlet_bool_mesh)[source]¶

Compute force equilibrium loss.

Parameters:

internal_forces_mesh (torch.Tensor(2d)) – Internal forces of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).
external_forces_mesh (torch.Tensor(2d)) – External forces of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).
reaction_forces_mesh (torch.Tensor(2d)) – Reaction forces (Dirichlet boundary conditions) of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).
dirichlet_bool_mesh (torch.Tensor(2d)) – Degrees of freedom of finite element mesh subject to Dirichlet boundary conditions. Stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim) where constrained degrees of freedom are labeled 1, otherwise 0.

Returns:

force_equilibrium_loss – Force equilibrium loss.

Return type:

float

force_equilibrium_loss_components_hist(internal_forces_mesh_hist, external_forces_mesh_hist, reaction_forces_mesh_hist, dirichlet_bc_mesh_hist)[source]¶

Compute force equilibrium loss components history (output purposes).

Parameters:

internal_forces_mesh_hist (torch.Tensor(3d)) – Internal forces history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
external_forces_mesh_hist (torch.Tensor(3d)) – External forces history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
reaction_forces_mesh_hist (torch.Tensor(3d)) – Reaction forces (Dirichlet boundary conditions) history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
dirichlet_bc_mesh_hist (torch.Tensor(3d)) – Dirichlet boundary constraints history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time). Encodes if each degree of freedom is free (assigned 0) or constrained (greater than 0) under Dirichlet boundary conditions. The encoding depends on the selected force equilibrium loss type.

Returns:

force_equilibrium_loss_components_hist – Force equilibrium loss components history stored as torch.Tensor(2d) of shape (1 + n_loss_comp, n_time).

Return type:

torch.Tensor(2d)

forward(sequential_mode='sequential_element')[source]¶

Forward propagation.

Parameters:

specimen_data (SpecimenNumericalData) – Specimen numerical data translated from experimental results.
specimen_material_state (StructureMaterialState) – FETorch structure material state.
sequential_mode ({'sequential_time', 'sequential_element', 'sequential_element_vmap}, default='sequential_element') –
‘sequential_time’ : Internal forces are computed in the standard way, processing each time step sequentially. Currently only available for inference.

’sequential_element’ : Internal forces are computed such that each element is processed sequentially (taking into account the corresponding deformation history). Available for both training and inference. Significantly limited with respect to memory costs.

’sequential_element_vmap’ : Similar to ‘sequential_element’ but leveraging vectorizing maps (significant improvement of processing time and memory efficiency). Available for both training and inference.

Returns:

force_equilibrium_hist_loss – Force equilibrium history loss.

Return type:

torch.Tensor(0d)

forward_sequential_element()[source]¶

Forward propagation (sequential element).

Returns:: force_equilibrium_hist_loss – Force equilibrium history loss.
Return type:: torch.Tensor(0d)

forward_sequential_time()[source]¶

Forward propagation (sequential time).

Returns:: force_equilibrium_hist_loss – Force equilibrium history loss.
Return type:: torch.Tensor(0d)

get_buffer(target)¶

Returns the buffer given by target if it exists, otherwise throws 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.

Parameters:: target (str) – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)
Returns:: The buffer referenced by target
Return type:: torch.Tensor
Raises:: AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_detached_model_parameters()[source]¶

Get model parameters (material models) detached of gradients.

Only collects parameters from material models with explicit learnable parameters.

Parameters names are prefixed by corresponding model label.

Returns:: model_parameters – Model parameters.
Return type:: dict

get_device()[source]¶

Get device on which torch.Tensor is allocated.

Parameters:

device_type ({'cpu', 'cuda'}) – Type of device on which torch.Tensor is allocated.
device (torch.device) – Device on which torch.Tensor is allocated.

get_extra_state()¶

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding 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:: Any extra state to store in the module’s state_dict
Return type:: object

get_material_models()[source]¶

Get material models.

Returns:: material_models – FETorch material constitutive models (key, str[int], item, ConstitutiveModel). Models are labeled from 1 to n_mat_model.
Return type:: dict

get_model_parameters_bounds()[source]¶

Get model parameters (material models) bounds.

Only collects parameters bounds from material models with explicit learnable parameters.

Parameters names are prefixed by corresponding model label.

Returns:: model_parameters_bounds – Model learnable parameters bounds. For each parameter (key, str), the corresponding bounds are stored as a tuple(lower_bound, upper_bound) (item, tuple).
Return type:: dict

get_parameter(target)¶

Returns the parameter given by target if it exists, otherwise throws 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.

Parameters:: target (str) – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)
Returns:: The Parameter referenced by target
Return type:: torch.nn.Parameter
Raises:: AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(target)¶

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

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.

Parameters:: target (str) – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
Returns:: The submodule referenced by target
Return type:: torch.nn.Module
Raises:: AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

half()¶

Casts all floating point parameters and buffers to half datatype.

Note

This method modifies the module in-place.

Returns:: self
Return type:: Module

ipu(device=None)¶

Moves 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.

Parameters:: device (int, optional) – if specified, all parameters will be copied to that device
Returns:: self
Return type:: Module

load_state_dict(state_dict, strict=True, assign=False)¶

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Warning

If assign is True the optimizer must be created after the call to load_state_dict.

Parameters:

state_dict (dict) – a dict containing parameters and persistent buffers.
strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True
assign (bool, optional) – whether to assign items in the state dictionary to their corresponding keys in the module instead of copying them inplace into the module’s current parameters and buffers. 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. Default: False

Returns:

missing_keys is a list of str containing the missing keys
unexpected_keys is a list of str containing the unexpected keys

Return type:

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

modules()¶

Returns an iterator over all modules in the network.

Yields:: Module – a module in the network
Return type:: Iterator[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.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)

named_buffers(prefix='', recurse=True, remove_duplicate=True)¶

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters:

prefix (str) – prefix to prepend to all buffer names.
recurse (bool, optional) – 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.
remove_duplicate (bool, optional) – whether to remove the duplicated buffers in the result. Defaults to True.

Yields:

(str, torch.Tensor) – Tuple containing the name and buffer

Return type:

Iterator[Tuple[str, Tensor]]

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>     if name in ['running_var']:
>>>         print(buf.size())

named_children()¶

Returns 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)

Return type:: Iterator[Tuple[str, Module]]

named_modules(memo=None, prefix='', remove_duplicate=True)¶

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters:

memo (Optional[Set[Module]]) – a memo to store the set of modules already added to the result
prefix (str) – a prefix that will be added to the name of the module
remove_duplicate (bool) – 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))

named_parameters(prefix='', recurse=True, remove_duplicate=True)¶

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters:

prefix (str) – prefix to prepend to all parameter names.
recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
remove_duplicate (bool, optional) – whether to remove the duplicated parameters in the result. Defaults to True.

Yields:

(str, Parameter) – Tuple containing the name and parameter

Return type:

Iterator[Tuple[str, Parameter]]

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>     if name in ['bias']:
>>>         print(param.size())

parameters(recurse=True)¶

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters:: recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
Yields:: Parameter – module parameter
Return type:: Iterator[Parameter]

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

recurrent_material_state_update(strain_formulation, problem_type, constitutive_model, strain_hist, time_hist)[source]¶

Material state update for any given recurrent constitutive model.

Parameters:

strain_formulation ({'infinitesimal', 'finite'}) – Problem strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
constitutive_model (ConstitutiveModel) – Recurrent material constitutive model.
strain_hist (torch.Tensor(3d)) – Strain tensor history stored as torch.Tensor(3d) of shape (n_dim, n_dim, n_time).
time_hist (torch.Tensor(1d)) – Discrete time history.

Returns:

state_variables_hist – Material constitutive model state variables history.

Return type:

list[dict]

register_backward_hook(hook)¶

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns:: a handle that can be used to remove the added hook by calling handle.remove()
Return type:: torch.utils.hooks.RemovableHandle

register_buffer(name, tensor, persistent=True)¶

Adds 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 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 state_dict.

Buffers can be accessed as attributes using given names.

Parameters:

name (str) – name of the buffer. The buffer can be accessed from this module using the given name
tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.
persistent (bool) – whether the buffer is part of this module’s state_dict.

Return type:

None

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook, *, prepend=False, with_kwargs=False, always_call=False)¶

Registers a forward hook on the module.

The hook will be called every time after 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 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

Parameters:

hook (Callable) – The user defined hook to be registered.
prepend (bool) – If True, the provided hook will be fired before all existing forward hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this torch.nn.modules.Module. Note that global forward hooks registered with register_module_forward_hook() will fire before all hooks registered by this method. Default: False
with_kwargs (bool) – If True, the hook will be passed the kwargs given to the forward function. Default: False
always_call (bool) – If True the hook will be run regardless of whether an exception is raised while calling the Module. Default: False

Returns:

a handle that can be used to remove the added hook by calling handle.remove()

Return type:

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook, *, prepend=False, with_kwargs=False)¶

Registers a forward pre-hook on the module.

The hook will be called every time before 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

Parameters:

hook (Callable) – The user defined hook to be registered.
prepend (bool) – If true, the provided hook will be fired before all existing forward_pre hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this torch.nn.modules.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook() will fire before all hooks registered by this method. Default: False
with_kwargs (bool) – If true, the hook will be passed the kwargs given to the forward function. Default: False

Returns:

a handle that can be used to remove the added hook by calling handle.remove()

Return type:

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook, prepend=False)¶

Registers 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 grad_input and 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 grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and 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.

Parameters:

hook (Callable) – The user-defined hook to be registered.
prepend (bool) – If true, the provided hook will be fired before all existing backward hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this torch.nn.modules.Module. Note that global backward hooks registered with register_module_full_backward_hook() will fire before all hooks registered by this method.

Returns:

a handle that can be used to remove the added hook by calling handle.remove()

Return type:

torch.utils.hooks.RemovableHandle

register_full_backward_pre_hook(hook, prepend=False)¶

Registers 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 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 grad_output in subsequent computations. Entries in 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.

Parameters:

hook (Callable) – The user-defined hook to be registered.
prepend (bool) – If true, the provided hook will be fired before all existing backward_pre hooks on this torch.nn.modules.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this torch.nn.modules.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook() will fire before all hooks registered by this method.

Returns:

a handle that can be used to remove the added hook by calling handle.remove()

Return type:

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)¶

Registers a post hook to be run after module’s 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 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:: a handle that can be used to remove the added hook by calling handle.remove()
Return type:: torch.utils.hooks.RemovableHandle

register_module(name, module)¶

Alias for add_module().

Return type:: None

register_parameter(name, param)¶

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters:

name (str) – name of the parameter. The parameter can be accessed from this module using the given name
param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

Return type:

None

register_state_dict_pre_hook(hook)¶: These hooks will be called with arguments: self, prefix, and keep_vars before calling state_dict on self. The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(requires_grad=True)¶

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ 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 locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters:: requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.
Returns:: self
Return type:: Module

set_device(device_type)[source]¶

Set device on which torch.Tensor is allocated.

Parameters:

device_type ({'cpu', 'cuda'}) – Type of device on which torch.Tensor is allocated.
device (torch.device) – Device on which torch.Tensor is allocated.

set_extra_state(state)¶

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters:: state (dict) – Extra state from the state_dict

set_fitted_force_data_scalers(force_minimum, force_maximum)[source]¶

Set fitted forces data scalers.

Parameters:

force_minimum (torch.Tensor(1d)) – Forces normalization minimum tensor stored as a torch.Tensor with shape (n_dim,).
force_maximum (torch.Tensor(1d)) – Forces normalization maximum tensor stored as a torch.Tensor with shape (n_dim,).

set_material_models_fitted_data_scalers(models_scaling_type, models_scaling_parameters)[source]¶

Set material constitutive models fitted data scalers.

Data scalers are only fitted for material models that support data normalization and for which the corresponding data scaling type and parameters are provided.

Parameters:

models_scaling_type (dict) – Type of data scaling (str, {‘min-max’, ‘mean-std’}) for each material model (key, str[int]). Models are labeled from 1 to n_mat_model. Min-Max scaling (‘min-max’) or standardization (‘mean-std’).
models_scaling_type – Features data scaling parameters (item, dict) for each material model (key, str[int]), stored as data scaling parameters (item, tuple[2]) for each features type (key, str). Models are labeled from 1 to n_mat_model. Each data scaling parameter is set as a torch.Tensor(1d) according to the corresponding number of features. For ‘min-max’ data scaling, the parameters are the ‘minimum’[0] and ‘maximum’[1] tensors, while for ‘mean-std’ data scaling the parameters are the ‘mean’[0] and ‘std’[1] tensors.

set_specimen_data(specimen_data, specimen_material_state, force_minimum=None, force_maximum=None, loss_scaling_factor=None, loss_time_weights=None)[source]¶

Set specimen data and material state.

Parameters:

specimen_data (SpecimenNumericalData) – Specimen numerical data translated from experimental results.
specimen_material_state (StructureMaterialState) – FETorch specimen material state.
force_minimum (torch.Tensor(1d), default=None) – Forces normalization minimum tensor stored as a torch.Tensor with shape (n_dim,). Only required if force normalization is set to True, otherwise ignored.
force_maximum (torch.Tensor(1d), default=None) – Forces normalization maximum tensor stored as a torch.Tensor with shape (n_dim,). Only required if force normalization is set to True, otherwise ignored.
loss_scaling_factor (torch.Tensor(0d), default=None) – Loss scaling factor. If provided, then loss is pre-multiplied by loss scaling factor.
loss_time_weights (torch.Tensor(1d), default=None) – Loss time weights stored as torch.Tensor(1d) of shape (n_time). If provided, then each discrete time loss contribution is pre-multiplied by corresponding weight. If None, time weights are set to 1.0.

share_memory()¶

See torch.Tensor.share_memory_()

Return type:: TypeVar(T, bound= Module)

state_dict(*args, destination=None, prefix='', keep_vars=False)¶

Returns 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.

Parameters:

destination (dict, optional) – 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.
prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns:

a dictionary containing a whole state of the module

Return type:

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

store_dirichlet_sets_reaction_hist(dirichlet_sets_reaction_hist, is_export_csv=True, is_plot=True)[source]¶

Store reaction forces history of Dirichlet boundary sets.

Parameters:

dirichlet_sets_reaction_hist (torch.Tensor(3d)) – Reaction forces history of Dirichlet boundary sets stored as torch.Tensor(3d) of shape (n_sets, 1, n_time).
is_export_csv (bool, default=True) – If True, then export the reaction force history of Dirichlet boundary sets to a ‘.csv’ file.
is_plot (bool, default=True) – If True, then plot the reaction force history of each Dirichlet boundary set.

store_force_equilibrium_loss_components_hist(force_equilibrium_loss_components_hist, is_plot=True)[source]¶

Store force equilibrium loss components history.

Parameters:

force_equilibrium_loss_components_hist (torch.Tensor(2d)) – Force equilibrium loss components history stored as torch.Tensor(2d) of shape (1 + n_loss_comp, n_time).
is_plot (bool, default=True) – If True, then plot force equilibrium loss components history.

classmethod store_tensor_comps(comps, tensor, device=None)[source]¶

Store strain/stress tensor components in array.

Parameters:

comps (tuple[str]) – Strain/Stress components order.
tensor (torch.Tensor(2d)) – Strain/Stress tensor.
device (torch.device, default=None) – Device on which torch.Tensor is allocated.

Returns:

comps_array – Strain/Stress components array.

Return type:

torch.Tensor(1d)

to(*args, **kwargs)¶

Moves and/or casts the parameters and buffers.

This can be called as

to(device=None, dtype=None, non_blocking=False)

to(dtype, non_blocking=False)

to(tensor, non_blocking=False)

to(memory_format=torch.channels_last)

Its signature is similar to torch.Tensor.to(), but only accepts floating point or complex dtypes. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When 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.

Parameters:

device (torch.device) – the desired device of the parameters and buffers in this module
dtype (torch.dtype) – the desired floating point or complex dtype of the parameters and buffers in this module
tensor (torch.Tensor) – Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
memory_format (torch.memory_format) – the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns:

self

Return type:

Module

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)

to_empty(*, device, recurse=True)¶

Moves the parameters and buffers to the specified device without copying storage.

Parameters:

device (torch.device) – The desired device of the parameters and buffers in this module.
recurse (bool) – Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns:

self

Return type:

Module

train(mode=True)¶

Sets 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. Dropout, BatchNorm, etc.

Parameters:: mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.
Returns:: self
Return type:: Module

type(dst_type)¶

Casts all parameters and buffers to dst_type.

Note

This method modifies the module in-place.

Parameters:: dst_type (type or string) – the desired type
Returns:: self
Return type:: Module

vassemble_internal_forces(elements_internal_forces, elements_mesh_indexes, n_node_mesh, n_dim)[source]¶

Assemble element internal forces into mesh counterpart.

Compatible with vectorized mapping.

Parameters:

elements_internal_forces (torch.Tensor(2d)) – Internal forces of finite elements nodes stored as torch.Tensor(2d) of shape (n_elem, n_node*n_dim).
elements_mesh_indexes (torch.Tensor(2d)) – Elements nodes degrees of freedom mesh indexes stored as torch.Tensor(2d) of shape (n_elem, n_node*n_dof_node).
n_node_mesh (int) – Number of nodes of finite element mesh.
n_dim (int) – Number of spatial dimensions.

Returns:

internal_forces_mesh – Internal forces of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).

Return type:

torch.Tensor(2d)

vbuild_internal_forces_mesh_hist(elements_internal_forces_hist, elements_mesh_indexes, n_node_mesh, n_dim)[source]¶

Build internal forces history of finite element mesh.

Compatible with vectorized mapping.

Parameters:

elements_internal_forces_hist (torch.Tensor(3d)) – Internal forces history of finite elements nodes stored as torch.Tensor(3d) of shape (n_elem, n_node*n_dim, n_time).
elements_mesh_indexes (torch.Tensor(2d)) – Elements nodes degrees of freedom mesh indexes stored as torch.Tensor(2d) of shape (n_elem, n_node*n_dof_node).
n_node_mesh (int) – Number of nodes of finite element mesh.
n_dim (int) – Number of spatial dimensions.

Returns:

internal_forces_mesh_hist – Internal forces history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).

Return type:

torch.Tensor(3d)

classmethod vbuild_tensor_from_comps(n_dim, comps, comps_array, device=None)[source]¶

Build strain/stress tensor from given components.

Compatible with vectorized mapping.

Parameters:

n_dim (int) – Problem number of spatial dimensions.
comps (tuple[str]) – Strain/Stress components order.
comps_array (torch.Tensor(1d)) – Strain/Stress components array.
device (torch.device, default=None) – Device on which torch.Tensor is allocated.

Returns:

tensor – Strain/Stress tensor.

Return type:

torch.Tensor(2d)

vcompute_element_internal_forces_hist(nodes_coords_hist, nodes_disps_hist, strain_formulation, problem_type, element_type, element_material, time_hist)[source]¶

Compute history of finite element internal forces.

Compatible with vectorized mapping.

Parameters:

nodes_coords_hist (torch.Tensor(3d)) – Coordinates history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
nodes_disps_hist (torch.Tensor(3d)) – Displacements history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
element_type (Element) – FETorch finite element.
element_material (ConstitutiveModel) – FETorch material constitutive model.
time_hist (torch.Tensor(1d)) – Discrete time history.

Returns:

element_internal_forces_hist (torch.Tensor(2d)) – Element internal forces history stored as torch.Tensor(2d) of shape (n_node*n_dim, n_time).
element_state_hist (torch.Tensor(3d)) – Element Gauss integration points strain and stress path history stored as torch.Tensor(3d) of shape (n_gauss, n_time, n_strain_comps + n_stress_comps).

vcompute_element_vol_grad_hist(nodes_coords_hist, nodes_disps_hist, strain_formulation, problem_type, element_type, time_hist)[source]¶

Compute history of finite element volumetric gradient operator.

Compatible with vectorized mapping.

Parameters:

nodes_coords_hist (torch.Tensor(3d)) – Coordinates history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
nodes_disps_hist (torch.Tensor(3d)) – Displacements history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
element_type (Element) – FETorch finite element.
time_hist (torch.Tensor(1d)) – Discrete time history.

Returns:

avg_vol_grad_operator_hist – Element average volumetric gradient operator history stored as torch.Tensor(3d) of shape (n_time, n_strain_comp, n_node*n_dim).

Return type:

torch.Tensor(3d)

vcompute_elements_internal_forces_hist(strain_formulation, problem_type, element_type, element_material, elements_coords_hist, elements_disps_hist, time_hist)[source]¶

Compute history of finite elements internal forces.

Compatible with vectorized mapping.

Vectorization constraints require that all the elements share the same element type (FETorch finite element) and share the same material constitutive model (FETorch material constitutive model).

Parameters:

strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
element_type (Element) – FETorch finite element.
element_material (ConstitutiveModel) – FETorch material constitutive model.
elements_coords_hist (torch.Tensor(4d)) – Coordinates history of finite elements nodes stored as torch.Tensor(4d) of shape (n_elem, n_node, n_dim, n_time).
elements_disps_hist (torch.Tensor(4d)) – Displacements history of finite elements nodes stored as torch.Tensor(4d) of shape (n_elem, n_node, n_dim, n_time).
time_hist (torch.Tensor(1d)) – Discrete time history.

Returns:

elements_internal_forces_hist (torch.Tensor(3d)) – Internal forces history of finite elements nodes stored as torch.Tensor(3d) of shape (n_elem, n_node*n_dim, n_time).
elements_state_hist (torch.Tensor(4d)) – Gauss integration points strain and stress path history of finite elements stored as torch.Tensor(4d) of shape (n_elem, n_gauss, n_time, n_strain_comps + n_stress_comps).

vcompute_local_dev_sym_gradient(grad_operator_sym, n_dim)[source]¶

Compute discrete deviatoric symmetric gradient operator.

Parameters:

grad_operator_sym (torch.Tensor(2d)) – Discrete symmetric gradient operator.
n_dim (int) – Number of spatial dimensions.

Returns:

dev_grad_operator_sym – Discrete deviatoric symmetric gradient operator.

Return type:

torch.Tensor(2d)

vcompute_local_gradient(nodes_coords, local_coords, comp_order, element_type, is_symmetric=True)[source]¶

Compute discrete gradient operator at given local point of element.

Compatible with vectorized mapping.

Parameters:

nodes_coords (torch.Tensor(2d)) – Nodes coordinates stored as torch.Tensor(2d) of shape (n_node, n_dof_node).
local_coords (torch.Tensor(1d)) – Local coordinates of point where strain is computed.
comp_order (tuple) – Strain/Stress components order associated to matricial form.
element_type (Element) – FETorch finite element.
is_symmetric (bool, default=True) – If True, then compute discrete symmetric gradient operator. Otherwise, compute non-symmetric discrete gradient operator.

Returns:

jacobian_det (torch.Tensor(0d)) – Determinant of element Jacobian at given local coordinates.
grad_operator (torch.Tensor(2d)) – Discrete gradient operator evaluated at given local coordinates.

vcompute_local_internal_forces(stress_vmf, grad_operator_sym, jacobian_det, weight)[source]¶

Compute local integration point internal forces contribution.

Compatible with vectorized mapping.

Internal forces are computed in the spatial configuration, i.e., based on the discrete symmetric gradient operator and the Cauchy stress tensor.

Parameters:

stress_vmf (torch.Tensor(1d)) – Cauchy stress tensor stored in Voigt matricial form.
grad_operator_sym (torch.Tensor(2d)) – Discrete symmetric gradient operator evaluated at given local coordinates.
jacobian_det (torch.Tensor(0d)) – Determinant of element jacobian evaluated at given local coordinates.
weight (torch.Tensor(0d)) – Local integration point weight.

Returns:

internal_forces – Integration point contribution to element internal forces.

Return type:

torch.Tensor(1d)

vcompute_local_internal_forces_hist(local_coords, weight, strain_formulation, problem_type, element_type, nodes_coords_hist, nodes_disps_hist, time_hist, element_material, is_volumetric_bar=False, avg_vol_grad_operator_hist=None)[source]¶

Compute local integration point internal force contribution history.

Compatible with vectorized mapping.

Parameters:

local_coords (torch.Tensor(1d)) – Local integration point coordinates.
weight (torch.Tensor(0d)) – Local integration point weight.
strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
element_type (Element) – FETorch finite element.
nodes_coords_hist (torch.Tensor(3d)) – Coordinates history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
nodes_disps_hist (torch.Tensor(3d)) – Displacements history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
time_hist (torch.Tensor(1d)) – Discrete time history.
element_material (ConstitutiveModel) – FETorch material constitutive model.
is_volumetric_bar (bool, default=False) – If True, then use volumetric strain averaging formulation (e.g., B-bar formulation under infinitesimal strains).
avg_vol_grad_operator_hist (torch.Tensor(3d), default=None) – Element average volumetric gradient operator history stored as torch.Tensor(3d) of shape (n_time, n_strain_comp, n_node*n_dim). Required only if volumetric strain averaging formulation is used.

Returns:

local_internal_forces_hist (torch.Tensor(2d)) – Local integration point contribution history to finite element internal forces stored as torch.Tensor(2d) of shape (n_node*n_dim, n_time).
local_state_variables_hist (torch.Tensor(2d)) – Local integration point strain and stress path history stored as torch.Tensor(2d) of shape (n_time, n_strain_comps + n_stress_comps).

vcompute_local_strain(nodes_coords, nodes_disps, local_coords, strain_formulation, n_dim, comp_order, element_type)[source]¶

Compute strain tensor at given local point of element.

Compatible with vectorized mapping.

Parameters:

nodes_coords (torch.Tensor(2d)) – Nodes coordinates stored as torch.Tensor(2d) of shape (n_node, n_dof_node).
nodes_disps (torch.Tensor(2d)) – Nodes displacements stored as torch.Tensor(2d) of shape (n_node, n_dof_node).
local_coords (torch.Tensor(1d)) – Local coordinates of point where strain is computed.
strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
n_dim (int) – Number of spatial dimensions.
comp_order (tuple) – Strain/Stress components order associated to matricial form.
element_type (Element) – FETorch finite element.

Returns:

strain (torch.Tensor(2d)) – Strain tensor at given local coordinates.
jacobian_det (torch.Tensor(0d)) – Determinant of element Jacobian at given local coordinates.
grad_operator_sym (torch.Tensor(2d)) – Discrete symmetric gradient operator evaluated at given local coordinates.

vcompute_local_strain_vbar(nodes_coords, nodes_disps, avg_vol_grad_operator, local_coords, strain_formulation, n_dim, comp_order, element_type)[source]¶

Compute strain tensor at given local point of element.

The strain tensor is computed using a volumetric strain averaging formulation (e.g., B-bar formulation under infinitesimal strains).

Compatible with vectorized mapping.

Parameters:

nodes_coords (torch.Tensor(2d)) – Nodes coordinates stored as torch.Tensor(2d) of shape (n_node, n_dof_node).
nodes_disps (torch.Tensor(2d)) – Nodes displacements stored as torch.Tensor(2d) of shape (n_node, n_dof_node).
avg_vol_grad_operator (torch.Tensor(2d)) – Element average volumetric gradient operator.
local_coords (torch.Tensor(1d)) – Local coordinates of point where strain is computed.
strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
n_dim (int) – Number of spatial dimensions.
comp_order (tuple) – Strain/Stress components order associated to matricial form.
element_type (Element) – FETorch finite element.

Returns:

strain (torch.Tensor(2d)) – Strain tensor at given local coordinates.
vbar_grad_operator_sym (torch.Tensor(2d)) – Modified discrete symmetric gradient operator evaluated at given local coordinates.

vcompute_local_vol_grad_operator_hist(local_coords, weight, strain_formulation, problem_type, element_type, nodes_coords_hist, nodes_disps_hist, time_hist)[source]¶

Compute local integration point gradient contribution history.

Compatible with vectorized mapping.

Parameters:

local_coords (torch.Tensor(1d)) – Local integration point coordinates.
weight (torch.Tensor(0d)) – Local integration point weight.
strain_formulation ({'infinitesimal', 'finite'}) – Strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
element_type (Element) – FETorch finite element.
nodes_coords_hist (torch.Tensor(3d)) – Coordinates history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
nodes_disps_hist (torch.Tensor(3d)) – Displacements history of finite element nodes stored as torch.Tensor(3d) of shape (n_node, n_dim, n_time).
time_hist (torch.Tensor(1d)) – Discrete time history.

Returns:

vol_grad_operator_hist (torch.Tensor(3d)) – Local integration point contribution history to finite element average volumetric gradient operator history stored as torch.Tensor(3d) of shape (n_time, n_strain_comp, n_node*n_dim).
vol_hist (torch.Tensor(1d)) – Local integration point contribution history to finite element volume history stored as torch.Tensor(1d) of shape (n_time,).

vcompute_local_vol_sym_gradient(grad_operator_sym, n_dim)[source]¶

Compute discrete volumetric symmetric gradient operator.

Parameters:

grad_operator_sym (torch.Tensor(2d)) – Discrete symmetric gradient operator.
n_dim (int) – Number of spatial dimensions.

Returns:

vol_grad_operator_sym – Discrete volumetric symmetric gradient operator.

Return type:

torch.Tensor(2d)

vforce_equilibrium_hist_loss(internal_forces_mesh_hist, external_forces_mesh_hist, reaction_forces_mesh_hist, dirichlet_bc_mesh_hist)[source]¶

Compute force equilibrium history loss.

Compatible with vectorized mapping.

Parameters:

internal_forces_mesh_hist (torch.Tensor(3d)) – Internal forces history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
external_forces_mesh_hist (torch.Tensor(3d)) – External forces history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
reaction_forces_mesh_hist (torch.Tensor(3d)) – Reaction forces (Dirichlet boundary conditions) history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time).
dirichlet_bc_mesh_hist (torch.Tensor(3d)) – Dirichlet boundary constraints history of finite element mesh nodes stored as torch.Tensor(3d) of shape (n_node_mesh, n_dim, n_time). Encodes if each degree of freedom is free (assigned 0) or constrained (greater than 0) under Dirichlet boundary conditions. The encoding depends on the selected force equilibrium loss type.

Returns:

force_equilibrium_hist_loss – Force equilibrium history loss.

Return type:

torch.Tensor(0d)

vforce_equilibrium_loss(internal_forces_mesh, external_forces_mesh, reaction_forces_mesh, dirichlet_bc_mesh)[source]¶

Compute force equilibrium loss.

Compatible with vectorized mapping.

Parameters:

internal_forces_mesh (torch.Tensor(2d)) – Internal forces of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).
external_forces_mesh (torch.Tensor(2d)) – External forces of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).
reaction_forces_mesh (torch.Tensor(2d)) – Reaction forces (Dirichlet boundary conditions) of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim).
dirichlet_bc_mesh (torch.Tensor(2d)) – Dirichlet boundary constraints of finite element mesh nodes stored as torch.Tensor(2d) of shape (n_node_mesh, n_dim). Encodes if each degree of freedom is free (assigned 0) or constrained (greater than 0) under Dirichlet boundary conditions. The encoding depends on the selected force equilibrium loss type.

Returns:

force_equilibrium_loss – Force equilibrium loss.

Return type:

torch.Tensor(0d)

vforward_sequential_element()[source]¶

Forward propagation (sequential element).

Compatible with vectorized mapping.

Returns:: force_equilibrium_hist_loss – Force equilibrium history loss.
Return type:: torch.Tensor(0d)

vrecurrent_material_state_update(strain_formulation, problem_type, constitutive_model, strain_hist, time_hist)[source]¶

Material state update for recurrent constitutive model.

Compatible with vectorized mapping.

Parameters:

strain_formulation ({'infinitesimal', 'finite'}) – Problem strain formulation.
problem_type (int) – Problem type: 2D plane strain (1), 2D plane stress (2), 2D axisymmetric (3) and 3D (4).
constitutive_model (ConstitutiveModel) – Recurrent material constitutive model.
strain_hist (torch.Tensor(3d)) – Strain tensor history stored as torch.Tensor(3d) of shape (n_time, n_dim, n_dim).
time_hist (torch.Tensor(1d)) – Discrete time history.

Returns:

state_variables_hist – Strain and stress path history stored as torch.Tensor(2d) of shape (n_time, n_strain_comps + n_stress_comps).

Return type:

torch.Tensor(2d)

classmethod vstore_tensor_comps(comps, tensor, device=None)[source]¶

Store strain/stress tensor components in array.

Compatible with vectorized mapping.

Parameters:

comps (tuple[str]) – Strain/Stress components order.
tensor (torch.Tensor(2d)) – Strain/Stress tensor.
device (torch.device, default=None) – Device on which torch.Tensor is allocated.

Returns:

comps_array – Strain/Stress components array.

Return type:

torch.Tensor(1d)

xpu(device=None)¶

Moves 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.

Parameters:: device (int, optional) – if specified, all parameters will be copied to that device
Returns:: self
Return type:: Module

zero_grad(set_to_none=True)¶

Resets gradients of all model parameters. See similar function under torch.optim.Optimizer for more context.

Parameters:: set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.
Return type:: None