deepmd.pt.model.descriptor.se_atten

Classes

`DescrptBlockSeAtten`	The building block of descriptor.
`NeighborGatedAttention`
`NeighborGatedAttentionLayer`
`GatedAttentionLayer`

Module Contents

class deepmd.pt.model.descriptor.se_atten.DescrptBlockSeAtten(rcut: float, rcut_smth: float, sel: List[int] | int, ntypes: int, neuron: list = [25, 50, 100], axis_neuron: int = 16, tebd_dim: int = 8, tebd_input_mode: str = 'concat', set_davg_zero: bool = True, attn: int = 128, attn_layer: int = 2, attn_dotr: bool = True, attn_mask: bool = False, activation_function='tanh', precision: str = 'float64', resnet_dt: bool = False, scaling_factor=1.0, normalize=True, temperature=None, smooth: bool = True, type_one_side: bool = False, exclude_types: List[Tuple[int, int]] = [], env_protection: float = 0.0, trainable_ln: bool = True, ln_eps: float | None = 1e-05, seed: int | List[int] | None = None, type: str | None = None, old_impl: bool = False)[source]

Bases: deepmd.pt.model.descriptor.descriptor.DescriptorBlock

The building block of descriptor. Given the input descriptor, provide with the atomic coordinates, atomic types and neighbor list, calculate the new descriptor.

rcut[source]

rcut_smth[source]

neuron[source]

filter_neuron[source]

axis_neuron[source]

tebd_dim[source]

tebd_input_mode[source]

set_davg_zero[source]

attn_dim[source]

attn_layer[source]

attn_dotr[source]

attn_mask[source]

activation_function[source]

precision[source]

prec[source]

resnet_dt[source]

scaling_factor[source]

normalize[source]

temperature[source]

smooth[source]

type_one_side[source]

env_protection[source]

trainable_ln[source]

seed[source]

ln_eps[source]

old_impl[source]

ntypes[source]

sel[source]

sec[source]

split_sel[source]

nnei[source]

ndescrpt[source]

wanted_shape[source]

mean[source]

stddev[source]

tebd_dim_input[source]

filter_layers_old = None[source]

filter_layers = None[source]

filter_layers_strip = None[source]

stats = None[source]

get_rcut() → float[source]: Returns the cut-off radius.

get_rcut_smth() → float[source]: Returns the radius where the neighbor information starts to smoothly decay to 0.

get_nsel() → int[source]: Returns the number of selected atoms in the cut-off radius.

get_sel() → List[int][source]: Returns the number of selected atoms for each type.

get_ntypes() → int[source]: Returns the number of element types.

get_dim_in() → int[source]: Returns the input dimension.

get_dim_out() → int[source]: Returns the output dimension.

get_dim_emb() → int[source]: Returns the output dimension of embedding.

__setitem__(key, value)[source]

__getitem__(key)[source]

mixed_types() → bool[source]

If true, the discriptor 1. assumes total number of atoms aligned across frames; 2. requires a neighbor list that does not distinguish different atomic types.

If false, the discriptor 1. assumes total number of atoms of each atom type aligned across frames; 2. requires a neighbor list that distinguishes different atomic types.

get_env_protection() → float[source]: Returns the protection of building environment matrix.

property dim_out[source]: Returns the output dimension of this descriptor.

property dim_in[source]: Returns the atomic input dimension of this descriptor.

property dim_emb[source]: Returns the output dimension of embedding.

compute_input_stats(merged: Callable[[], List[dict]] | List[dict], path: deepmd.utils.path.DPPath | None = None)[source]

Compute the input statistics (e.g. mean and stddev) for the descriptors from packed data.

Parameters:

mergedUnion[Callable[[], List[dict]], List[dict]]

List[dict]: A list of data samples from various data systems.
Each element, merged[i], is a data dictionary containing keys: torch.Tensor originating from the i-th data system.
Callable[[], List[dict]]: A lazy function that returns data samples in the above format
only when needed. Since the sampling process can be slow and memory-intensive, the lazy function helps by only sampling once.

pathOptional[DPPath]

The path to the stat file.

get_stats() → Dict[str, deepmd.utils.env_mat_stat.StatItem][source]: Get the statistics of the descriptor.

reinit_exclude(exclude_types: List[Tuple[int, int]] = [])[source]

forward(nlist: torch.Tensor, extended_coord: torch.Tensor, extended_atype: torch.Tensor, extended_atype_embd: torch.Tensor | None = None, mapping: torch.Tensor | None = None)[source]

Compute the descriptor.

Parameters:

nlist: The neighbor list. shape: nf x nloc x nnei
extended_coord: The extended coordinates of atoms. shape: nf x (nallx3)
extended_atype: The extended aotm types. shape: nf x nall x nt
extended_atype_embd: The extended type embedding of atoms. shape: nf x nall
mapping: The index mapping, not required by this descriptor.

Returns:

result: The descriptor. shape: nf x nloc x (ng x axis_neuron)
g2: The rotationally invariant pair-partical representation. shape: nf x nloc x nnei x ng
h2: The rotationally equivariant pair-partical representation. shape: nf x nloc x nnei x 3
gr: The rotationally equivariant and permutationally invariant single particle representation. shape: nf x nloc x ng x 3
sw: The smooth switch function. shape: nf x nloc x nnei

has_message_passing() → bool[source]: Returns whether the descriptor block has message passing.

need_sorted_nlist_for_lower() → bool[source]: Returns whether the descriptor block needs sorted nlist when using forward_lower.

class deepmd.pt.model.descriptor.se_atten.NeighborGatedAttention(layer_num: int, nnei: int, embed_dim: int, hidden_dim: int, dotr: bool = False, do_mask: bool = False, scaling_factor: float = 1.0, normalize: bool = True, temperature: float | None = None, trainable_ln: bool = True, ln_eps: float = 1e-05, smooth: bool = True, precision: str = DEFAULT_PRECISION, seed: int | List[int] | None = None)[source]

Bases: torch.nn.Module

layer_num[source]

nnei[source]

embed_dim[source]

hidden_dim[source]

dotr[source]

do_mask[source]

scaling_factor[source]

normalize[source]

temperature[source]

trainable_ln[source]

ln_eps[source]

smooth[source]

precision[source]

seed[source]

network_type[source]

attention_layers = [][source]

forward(input_G, nei_mask, input_r: torch.Tensor | None = None, sw: torch.Tensor | None = None)[source]

Compute the multi-layer gated self-attention.

Parameters:

input_G: inputs with shape: (nf x nloc) x nnei x embed_dim.
nei_mask: neighbor mask, with paddings being 0. shape: (nf x nloc) x nnei.
input_r: normalized radial. shape: (nf x nloc) x nnei x 3.
sw: The smooth switch function. shape: nf x nloc x nnei

__getitem__(key)[source]

__setitem__(key, value)[source]

serialize() → dict[source]

Serialize the networks to a dict.

Returns:

dict: The serialized networks.

classmethod deserialize(data: dict) → NeighborGatedAttention[source]

Deserialize the networks from a dict.

Parameters:

datadict: The dict to deserialize from.

class deepmd.pt.model.descriptor.se_atten.NeighborGatedAttentionLayer(nnei: int, embed_dim: int, hidden_dim: int, dotr: bool = False, do_mask: bool = False, scaling_factor: float = 1.0, normalize: bool = True, temperature: float | None = None, smooth: bool = True, trainable_ln: bool = True, ln_eps: float = 1e-05, precision: str = DEFAULT_PRECISION, seed: int | List[int] | None = None)[source]

Bases: torch.nn.Module

nnei[source]

embed_dim[source]

hidden_dim[source]

dotr[source]

do_mask[source]

scaling_factor[source]

normalize[source]

temperature[source]

precision[source]

trainable_ln[source]

ln_eps[source]

seed[source]

attention_layer[source]

attn_layer_norm[source]

forward(x, nei_mask, input_r: torch.Tensor | None = None, sw: torch.Tensor | None = None)[source]

serialize() → dict[source]

Serialize the networks to a dict.

Returns:

dict: The serialized networks.

classmethod deserialize(data: dict) → NeighborGatedAttentionLayer[source]

Deserialize the networks from a dict.

Parameters:

datadict: The dict to deserialize from.

class deepmd.pt.model.descriptor.se_atten.GatedAttentionLayer(nnei: int, embed_dim: int, hidden_dim: int, num_heads: int = 1, dotr: bool = False, do_mask: bool = False, scaling_factor: float = 1.0, normalize: bool = True, temperature: float | None = None, bias: bool = True, smooth: bool = True, precision: str = DEFAULT_PRECISION, seed: int | List[int] | None = None)[source]

Bases: torch.nn.Module

nnei[source]

embed_dim[source]

hidden_dim[source]

num_heads[source]

head_dim[source]

dotr[source]

do_mask[source]

bias[source]

smooth[source]

scaling_factor[source]

temperature[source]

precision[source]

seed[source]

scaling[source]

normalize[source]

in_proj[source]

out_proj[source]

forward(query, nei_mask, input_r: torch.Tensor | None = None, sw: torch.Tensor | None = None, attnw_shift: float = 20.0)[source]

Compute the multi-head gated self-attention.

Parameters:

query: inputs with shape: (nf x nloc) x nnei x embed_dim.
nei_mask: neighbor mask, with paddings being 0. shape: (nf x nloc) x nnei.
input_r: normalized radial. shape: (nf x nloc) x nnei x 3.
sw: The smooth switch function. shape: (nf x nloc) x nnei
attnw_shiftfloat: The attention weight shift to preserve smoothness when doing padding before softmax.

serialize() → dict[source]

Serialize the networks to a dict.

Returns:

dict: The serialized networks.

classmethod deserialize(data: dict) → GatedAttentionLayer[source]

Deserialize the networks from a dict.

Parameters:

datadict: The dict to deserialize from.