import keras as ks
from kgcnn.layers.scale import get as get_scaler
from ._model import model_disjoint
from kgcnn.layers.modules import Input
from kgcnn.models.casting import (template_cast_output, template_cast_list_input,
template_cast_list_input_docs, template_cast_output_docs)
from kgcnn.models.utils import update_model_kwargs
from keras.backend import backend as backend_to_use
# To be updated if model is changed in a significant way.
__model_version__ = "2023-12-04"
# Supported backends
__kgcnn_model_backend_supported__ = ["tensorflow", "torch", "jax"]
if backend_to_use() not in __kgcnn_model_backend_supported__:
raise NotImplementedError("Backend '%s' for model 'EGNN' is not supported." % backend_to_use())
# Implementation of EGNN in `keras` from paper:
# E(n) Equivariant Graph Neural Networks
# by Victor Garcia Satorras, Emiel Hoogeboom, Max Welling (2021)
# https://arxiv.org/abs/2102.09844
model_default = {
"name": "EGNN",
"inputs": [
{"shape": (None,), "name": "node_number", "dtype": "int64"},
{"shape": (None, 3), "name": "node_coordinates", "dtype": "float32"},
{"shape": (None, 10), "name": "edge_attributes", "dtype": "float32"},
{"shape": (None, 2), "name": "edge_indices", "dtype": "int64"},
{"shape": (), "name": "total_nodes", "dtype": "int64"},
{"shape": (), "name": "total_edges", "dtype": "int64"},
],
"input_tensor_type": "padded",
"cast_disjoint_kwargs": {},
"input_embedding": None,
"input_node_embedding": {"input_dim": 95, "output_dim": 64},
"input_edge_embedding": {"input_dim": 95, "output_dim": 64},
"depth": 4,
"node_mlp_initialize": None,
"euclidean_norm_kwargs": {"keepdims": True, "axis": -1},
"use_edge_attributes": True,
"edge_mlp_kwargs": {"units": [64, 64], "activation": ["swish", "linear"]},
"edge_attention_kwargs": None, # {"units: 1", "activation": "sigmoid"}
"use_normalized_difference": False,
"expand_distance_kwargs": None,
"coord_mlp_kwargs": {"units": [64, 1], "activation": ["swish", "linear"]}, # option: "tanh" at the end.
"pooling_coord_kwargs": {"pooling_method": "mean"},
"pooling_edge_kwargs": {"pooling_method": "sum"},
"node_normalize_kwargs": None,
"use_node_attributes": False,
"node_mlp_kwargs": {"units": [64, 64], "activation": ["swish", "linear"]},
"use_skip": True,
"verbose": 10,
"node_decoder_kwargs": None,
"node_pooling_kwargs": {"pooling_method": "sum"},
"output_embedding": "graph",
"output_to_tensor": None, # deprecated
"output_tensor_type": "padded",
"output_mlp": {"use_bias": [True, True], "units": [64, 1],
"activation": ["swish", "linear"]},
"output_scaling": None,
}
[docs]@update_model_kwargs(model_default, update_recursive=0, deprecated=["input_embedding", "output_to_tensor"])
def make_model(name: str = None,
inputs: list = None,
input_tensor_type: str = None,
cast_disjoint_kwargs: dict = None,
input_embedding: dict = None, # noqa
input_node_embedding: dict = None,
input_edge_embedding: dict = None,
depth: int = None,
euclidean_norm_kwargs: dict = None,
node_mlp_initialize: dict = None,
use_edge_attributes: bool = None,
edge_mlp_kwargs: dict = None,
edge_attention_kwargs: dict = None,
use_normalized_difference: bool = None,
expand_distance_kwargs: dict = None,
coord_mlp_kwargs: dict = None,
pooling_coord_kwargs: dict = None,
pooling_edge_kwargs: dict = None,
node_normalize_kwargs: dict = None,
use_node_attributes: bool = None,
node_mlp_kwargs: dict = None,
use_skip: bool = None,
verbose: int = None, # noqa
node_decoder_kwargs: dict = None,
node_pooling_kwargs: dict = None,
output_embedding: str = None,
output_to_tensor: bool = None, # noqa
output_mlp: dict = None,
output_tensor_type: str = None,
output_scaling: dict = None
):
r"""Make `EGNN <https://arxiv.org/abs/2102.09844>`__ graph network via functional API.
Default parameters can be found in :obj:`kgcnn.literature.EGNN.model_default`.
**Model inputs**:
Model uses the list template of inputs and standard output template.
The supported inputs are :obj:`[nodes, node_coordinates, edge_attributes, edge_indices, ...]`
with '...' indicating mask or ID tensors following the template below.
%s
**Model outputs**:
The standard output template:
%s
Args:
name (str): Name of the model. Default is "EGNN".
inputs (list): List of dictionaries unpacked in :obj:`tf.keras.layers.Input`. Order must match model definition.
cast_disjoint_kwargs (dict): Dictionary of arguments for casting layers if used.
input_tensor_type (str): Input type of graph tensor. Default is "padded".
input_embedding (dict): Deprecated in favour of input_node_embedding etc.
input_node_embedding (dict): Dictionary of arguments for nodes unpacked in :obj:`Embedding` layers.
input_edge_embedding (dict): Dictionary of arguments for edge unpacked in :obj:`Embedding` layers.
depth (int): Number of graph embedding units or depth of the network.
euclidean_norm_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`EuclideanNorm`.
node_mlp_initialize (dict): Dictionary of layer arguments unpacked in :obj:`GraphMLP` layer for start embedding.
use_edge_attributes (bool): Whether to use edge attributes including for example further edge information.
edge_mlp_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`GraphMLP` layer.
edge_attention_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`GraphMLP` layer.
use_normalized_difference (bool): Whether to use a normalized difference vector for nodes.
expand_distance_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`PositionEncodingBasisLayer`.
coord_mlp_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`GraphMLP` layer.
pooling_coord_kwargs (dict):
pooling_edge_kwargs (dict):
node_normalize_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`GraphLayerNormalization` layer.
use_node_attributes (bool): Whether to add node attributes before node MLP.
node_mlp_kwargs (dict):
use_skip (bool):
verbose (int): Level of verbosity.
node_decoder_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`MLP` layer after graph network.
node_pooling_kwargs (dict): Dictionary of layer arguments unpacked in :obj:`PoolingNodes` layers.
output_embedding (str): Main embedding task for graph network. Either "node", "edge" or "graph".
output_to_tensor (bool): Deprecated in favour of `output_tensor_type` .
output_tensor_type (str): Output type of graph tensors such as nodes or edges. Default is "padded".
output_mlp (dict): Dictionary of layer arguments unpacked in the final classification :obj:`MLP` layer block.
Defines number of model outputs and activation.
output_scaling (dict): Dictionary of layer arguments unpacked in scaling layers.
Returns:
:obj:`keras.models.Model`
"""
# Make input
model_inputs = [Input(**x) for x in inputs]
dj = template_cast_list_input(
model_inputs,
input_tensor_type=input_tensor_type,
cast_disjoint_kwargs=cast_disjoint_kwargs,
index_assignment=[None, None, None, 0],
mask_assignment=[0, 0, 1, 1]
)
n, x, ed, disjoint_indices, batch_id_node, batch_id_edge, node_id, edge_id, count_nodes, count_edges = dj
out = model_disjoint(
[n, x, ed, disjoint_indices, batch_id_node, batch_id_edge, count_nodes, count_edges],
use_node_embedding=("int" in inputs[0]['dtype']) if input_node_embedding is not None else False,
use_edge_embedding=("int" in inputs[2]['dtype']) if input_edge_embedding is not None else False,
input_node_embedding=input_node_embedding,
input_edge_embedding=input_edge_embedding,
depth=depth,
euclidean_norm_kwargs=euclidean_norm_kwargs,
node_mlp_initialize=node_mlp_initialize,
use_edge_attributes=use_edge_attributes,
edge_mlp_kwargs=edge_mlp_kwargs,
edge_attention_kwargs=edge_attention_kwargs,
use_normalized_difference=use_normalized_difference,
expand_distance_kwargs=expand_distance_kwargs,
coord_mlp_kwargs=coord_mlp_kwargs,
pooling_coord_kwargs=pooling_coord_kwargs,
pooling_edge_kwargs=pooling_edge_kwargs,
node_normalize_kwargs=node_normalize_kwargs,
use_node_attributes=use_node_attributes,
node_mlp_kwargs=node_mlp_kwargs,
use_skip=use_skip,
node_decoder_kwargs=node_decoder_kwargs,
node_pooling_kwargs=node_pooling_kwargs,
output_embedding=output_embedding,
output_mlp=output_mlp
)
if output_scaling is not None:
scaler = get_scaler(output_scaling["name"])(**output_scaling)
if scaler.extensive:
# Node information must be numbers, or we need an additional input.
out = scaler([out, n, batch_id_node])
else:
out = scaler(out)
# Output embedding choice
out = template_cast_output(
[out, batch_id_node, batch_id_edge, node_id, edge_id, count_nodes, count_edges],
output_embedding=output_embedding, output_tensor_type=output_tensor_type,
input_tensor_type=input_tensor_type, cast_disjoint_kwargs=cast_disjoint_kwargs,
)
model = ks.models.Model(inputs=model_inputs, outputs=out, name=name)
model.__kgcnn_model_version__ = __model_version__
if output_scaling is not None:
def set_scale(*args, **kwargs):
scaler.set_scale(*args, **kwargs)
setattr(model, "set_scale", set_scale)
return model
make_model.__doc__ = make_model.__doc__ % (template_cast_list_input_docs, template_cast_output_docs)