core.models.uma.escn_moe

core.models.uma.escn_moe#

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

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

Classes#

eSCNMDMoeBackbone

Base class for all neural network modules.

DatasetSpecificMoEWrapper

Base class for all neural network modules.

DatasetSpecificSingleHeadWrapper

Base class for all neural network modules.

Module Contents#

class core.models.uma.escn_moe.eSCNMDMoeBackbone(num_experts: int = 8, moe_dropout: float = 0.0, use_global_embedding: bool = False, use_composition_embedding: bool = False, moe_expert_coefficient_norm: str = 'softmax', act=torch.nn.SiLU, layers_moe=None, moe_layer_type: str = 'pytorch', moe_single: bool = False, moe_type: str = 'so2', **kwargs)#

Bases: fairchem.core.models.uma.escn_md.eSCNMDBackbone, fairchem.core.models.uma.nn.mole_utils.MOLEInterface

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing them to be nested 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) -> None:
        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 also have their parameters converted 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.

parent_kwargs#
num_experts#
merge_MOLE_model(data)#
set_MOLE_coefficients(atomic_numbers_full, batch_full, csd_mixed_emb)#
set_MOLE_sizes(nsystems, batch_full, edge_index)#
log_MOLE_stats()#
class core.models.uma.escn_moe.DatasetSpecificMoEWrapper(backbone, dataset_names, head_cls, wrap_property=True, head_kwargs=None)#

Bases: torch.nn.Module, fairchem.core.models.base.HeadInterface

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing them to be nested 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) -> None:
        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 also have their parameters converted 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.

regress_stress#
regress_forces#
wrap_property#
dataset_names#
dataset_name_to_exp#
head#
global_mole_tensors#
forward(data, emb: dict[str, torch.Tensor]) → dict[str, torch.Tensor]#

Head forward.

Parameters:

  • data (AtomicData) – 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.uma.escn_moe.DatasetSpecificSingleHeadWrapper(backbone, dataset_names, head_cls, wrap_property=True, head_kwargs=None)#

Bases: torch.nn.Module, fairchem.core.models.base.HeadInterface

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing them to be nested 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) -> None:
        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 also have their parameters converted 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.

regress_stress#
regress_forces#
wrap_property#
dataset_names#
head#
forward(data, emb: dict[str, torch.Tensor]) → dict[str, torch.Tensor]#

Head forward.

Parameters:

  • data (AtomicData) – 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]

Contents