core.models.base

Copyright (c) Meta, Inc. and its affiliates.

This source code is licensed under the MIT license found in the LICENSE file in the root directory of this source tree.

Classes

GraphData	Class to keep graph attributes nicely packaged.
GraphModelMixin	Mixin Model class implementing some general convenience properties and methods.
HeadInterface
BackboneInterface
HydraModel	Base class for all neural network modules.

Module Contents

class core.models.base.GraphData

Class to keep graph attributes nicely packaged.

edge_index: torch.Tensor

edge_distance: torch.Tensor

edge_distance_vec: torch.Tensor

cell_offsets: torch.Tensor

offset_distances: torch.Tensor

neighbors: torch.Tensor

batch_full: torch.Tensor

atomic_numbers_full: torch.Tensor

node_offset: int = 0

class core.models.base.GraphModelMixin

Mixin Model class implementing some general convenience properties and methods.

generate_graph(data, cutoff=None, max_neighbors=None, use_pbc=None, otf_graph=None, enforce_max_neighbors_strictly=None, use_pbc_single=False)

property num_params: int

no_weight_decay() → list

Returns a list of parameters with no weight decay.

class core.models.base.HeadInterface

property use_amp

abstract forward(data: torch_geometric.data.Batch, emb: dict[str, torch.Tensor]) → dict[str, torch.Tensor]

Head forward.

Parameters:

• data (DataBatch) – Atomic systems as input

• emb (dict[str->torch.Tensor]) – Embeddings of the input as generated by the backbone

Returns:

outputs – Return one or more targets generated by this head

Return type:

dict[str->torch.Tensor]

class core.models.base.BackboneInterface

abstract forward(data: torch_geometric.data.Batch) → dict[str, torch.Tensor]

Backbone forward.

Parameters:

data (DataBatch) – Atomic systems as input

Returns:

embedding – Return backbone embeddings for the given input

Return type:

dict[str->torch.Tensor]

class core.models.base.HydraModel(backbone: dict | None = None, heads: dict | None = None, finetune_config: dict | None = None, otf_graph: bool = True, pass_through_head_outputs: bool = False, freeze_backbone: bool = False)

Bases: torch.nn.Module, GraphModelMixin

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

Note

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

Variables:

training (bool) – Boolean represents whether this module is in training or evaluation mode.

device = None

otf_graph

pass_through_head_outputs

forward(data: torch_geometric.data.Batch)