tfep.nn.embeddings.radial.BehlerParrinelloRadialExpansion

class tfep.nn.embeddings.radial.BehlerParrinelloRadialExpansion(r_cutoff, means, stds, trainable_means=False, trainable_stds=False, force_zero_after_cutoff=True)

Bases: GaussianBasisExpansion

Expands distance into a soft one-hot encoded vector using a Gaussian basis with a cosine switching function.

This is a Gaussian radial basis expansion multiplied by a switching function similar to that used in Behler-Parrinello neural networks [1].

The means and bandwidths of the Gaussians can be specified as trainable parameters.

Parameters:

• r_cutoff (float) – The cutoff for the switching function in the same units used for the input distances in forward().

• means (torch.Tensor) – A tensor of shape (n_gaussians,) where means[i] is the center of the i-th Gaussian. The units must be the same used for the input distances in forward().

• stds (torch.Tensor) – A tensor of shape (n_gaussians,) where stds[i] is the standard deviation of the i-th Gaussian. The units must be the same used for the input distances in forward().

• trainable_means (bool, optional) – If True, the means are defined as parameters of the neural network and optimized during training.

• trainable_stds (bool, optional) – If True, the standard deviations are defined as parameters of the neural network and optimized during training.

• force_zero_after_cutoff (bool, optional) – If False, the function assumes that values after the cutoff are not provided and thus no element of the switching function needs to be explicitly set to 0.0. This can save a calculation if you have already removed distances greater than r_cutoff from the input.

References

[1] Behler J, Parrinello M. Generalized neural-network representation of

high-dimensional potential-energy surfaces. Physical review letters. 2007 Apr 2;98(14):146401.

__init__(r_cutoff, means, stds, trainable_means=False, trainable_stds=False, force_zero_after_cutoff=True)

Initialize internal Module state, shared by both nn.Module and ScriptModule.

Methods

__init__(r_cutoff, means, stds[, ...])	Initialize internal Module state, shared by both nn.Module and ScriptModule.
add_module(name, module)	Add a child module to the current module.
apply(fn)	Apply 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([recurse])	Return an iterator over module buffers.
children()	Return an iterator over immediate children modules.
compile(*args, **kwargs)	Compile this Module's forward using torch.compile().
cpu()	Move all model parameters and buffers to the CPU.
cuda([device])	Move all model parameters and buffers to the GPU.
double()	Casts all floating point parameters and buffers to double datatype.
eval()	Set the module in evaluation mode.
extra_repr()	Set the extra representation of the module.
float()	Casts all floating point parameters and buffers to float datatype.
forward(distances)	Expand a matrix of distances into a soft one-hot representation.
from_range(r_cutoff, n_gaussians, max_mean)	Create a basis of equidistant Gaussians in a given range.
get_buffer(target)	Return the buffer given by target if it exists, otherwise throw an error.
get_extra_state()	Return any extra state to include in the module's state_dict.
get_parameter(target)	Return the parameter given by target if it exists, otherwise throw an error.
get_submodule(target)	Return the submodule given by target if it exists, otherwise throw an error.
half()	Casts all floating point parameters and buffers to half datatype.
ipu([device])	Move all model parameters and buffers to the IPU.
load_state_dict(state_dict[, strict, assign])	Copy parameters and buffers from state_dict into this module and its descendants.
modules()	Return an iterator over all modules in the network.
mtia([device])	Move all model parameters and buffers to the MTIA.
named_buffers([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters([recurse])	Return an iterator over module parameters.
register_backward_hook(hook)	Register a backward hook on the module.
register_buffer(name, tensor[, persistent])	Add a buffer to the module.
register_forward_hook(hook, *[, prepend, ...])	Register a forward hook on the module.
register_forward_pre_hook(hook, *[, ...])	Register a forward pre-hook on the module.
register_full_backward_hook(hook[, prepend])	Register a backward hook on the module.
register_full_backward_pre_hook(hook[, prepend])	Register a backward pre-hook on the module.
register_load_state_dict_post_hook(hook)	Register a post-hook to be run after module's load_state_dict() is called.
register_load_state_dict_pre_hook(hook)	Register a pre-hook to be run before module's load_state_dict() is called.
register_module(name, module)	Alias for add_module().
register_parameter(name, param)	Add a parameter to the module.
register_state_dict_post_hook(hook)	Register a post-hook for the state_dict() method.
register_state_dict_pre_hook(hook)	Register a pre-hook for the state_dict() method.
requires_grad_([requires_grad])	Change if autograd should record operations on parameters in this module.
set_extra_state(state)	Set extra state contained in the loaded state_dict.
set_submodule(target, module)	Set the submodule given by target if it exists, otherwise throw an error.
share_memory()	See torch.Tensor.share_memory_().
state_dict(*args[, destination, prefix, ...])	Return a dictionary containing references to the whole state of the module.
to(*args, **kwargs)	Move and/or cast the parameters and buffers.
to_empty(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
train([mode])	Set the module in training mode.
type(dst_type)	Casts all parameters and buffers to dst_type.
xpu([device])	Move all model parameters and buffers to the XPU.
zero_grad([set_to_none])	Reset gradients of all model parameters.

Attributes

T_destination
call_super_init
dump_patches
training

forward(distances)

Expand a matrix of distances into a soft one-hot representation.

Parameters:

distances (torch.Tensor) – Distance matrix of shape [batch_size, n_atoms, n_atoms] or [batch_size, n_atoms, n_atoms, 1] where distances[b, i, j] represent the distance between atoms i and j for the b-th batch.

Returns:

encoding – A matrix of shape [batch_size, n_atoms, n_atoms, n_gaussians] where n_gaussians is the number of Gaussian basis function used to expand the distance.

Return type:

torch.Tensor

classmethod from_range(r_cutoff, n_gaussians, max_mean, min_mean=0.0, relative_std=3.0, trainable_means=False, trainable_stds=False, force_zero_after_cutoff=True)

Create a basis of equidistant Gaussians in a given range.

By default, standard deviations are set equal to three times the displacement between two consecutive gaussians.

Parameters:

• r_cutoff (float) – The cutoff for the switching function in the same units used for the input distances in forward().

• n_gaussians (int) – The number of equidistant Gaussians.

• max_mean (float) – The largest mean of the Gaussian in the same units used for the input distances in forward().

• min_mean (float, optional) – The smallest mean of the Gaussian in the same units used for the input distances in forward().

• relative_std (float, optional) – The standard deviation of each Gaussian relative to the displacement between two means. I.e., the std of the Gaussians will be set to relative_std * (means[i] - means[i-1]).

• trainable_means (bool, optional) – If True, the means are defined as parameters of the neural network and optimized during training.

• trainable_stds (bool, optional) – If True, the standard deviations are defined as parameters of the neural network and optimized during training.

• force_zero_after_cutoff (bool, optional) – If False, the function assumes that values after the cutoff are not provided and thus no element of the switching function needs to be explicitly set to 0.0. This can save a calculation if you have already removed distances greater than r_cutoff from the input.