# Copyright 2022 Huawei Technologies Co., Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ============================================================================
"""CFConv Layer"""
import mindspore as ms
from mindspore_gl import Graph
from .. import GNNCell
class ShiftedSoftplus(ms.nn.Cell):
"""Shifted soft plus."""
def __init__(self, shift=2.):
super().__init__()
self.shift = ms.Tensor([shift], ms.float32)
self.softplus = ms.ops.Softplus()
# pylint: disable=arguments-differ
def construct(self, x):
"""
Construct function for ShiftedSoftplus.
"""
return self.softplus(x) - ms.ops.Log()(self.shift)
[文档]class CFConv(GNNCell):
r"""
CFConv in SchNet.
From the paper `SchNet: A continuous-filter convolutional neural network for modeling quantum
interactions <https://arxiv.org/abs/1706.08566>`_ .
It combines node and edge features in messaging and updates node representations.
.. math::
h_i^{(l+1)} = \sum_{j\in \mathcal{N}(i)} h_j^{l} \circ W^{(l)}e_ij
Where :math:`SPP` represents:
.. math::
\text{SSP}(x) = \frac{1}{\beta} * \log(1 + \exp(\beta * x)) - \log(\text{shift})
Args:
node_feat_size (int): Node feature size.
edge_feat_size (int): Edge feature size.
hidden_size (int): Hidden layer size.
out_size (int): Output classes size.
Inputs:
- **x** (Tensor): The input node features. The shape is :math:`(N,*)` where :math:`N` is the number of nodes,
and :math:`*` could be of any shape.
- **edge_feats** (Tensor): The input edge features. The shape is :math:`(M,*)` where :math:`M` is the number of
edges, and :math:`*` could be of any shape.
- **g** (Graph): The input graph.
Outputs:
- Tensor, output node features. The shape is :math:`(N, out\_size)`.
Raises:
TypeError: If 'node_feat_size' is not a positive int.
TypeError: If 'edge_feat_size' is not a positive int.
TypeError: If 'hidden_size' is not a positive int.
TypeError: If 'out_size' is not a positive int.
Supported Platforms:
``Ascend`` ``GPU``
Examples:
>>> import mindspore as ms
>>> from mindspore_gl.nn import CFConv
>>> from mindspore_gl import GraphField
>>> n_nodes = 4
>>> n_edges = 8
>>> feat_size = 16
>>> src_idx = ms.Tensor([0, 0, 0, 1, 1, 1, 2, 3], ms.int32)
>>> dst_idx = ms.Tensor([0, 1, 3, 1, 2, 3, 3, 2], ms.int32)
>>> ones = ms.ops.Ones()
>>> nodes_feat = ones((n_nodes, feat_size), ms.float32)
>>> edges_feat = ones((n_edges, feat_size), ms.float32)
>>> graph_field = GraphField(src_idx, dst_idx, n_nodes, n_edges)
>>> hidden_size = 8
>>> out_size = 4
>>> conv = CFConv(feat_size, feat_size, hidden_size, out_size)
>>> ret = conv(nodes_feat, edges_feat, *graph_field.get_graph())
>>> print(ret.shape)
(4, 4)
"""
def __init__(self,
node_feat_size: int,
edge_feat_size: int,
hidden_size: int,
out_size: int
):
super().__init__()
assert isinstance(node_feat_size, int) and node_feat_size > 0, "node_feat_size must be positive int"
assert isinstance(edge_feat_size, int) and edge_feat_size > 0, "edge_feat_size must be positive int"
assert isinstance(hidden_size, int) and hidden_size > 0, "hidden_size must be positive int"
assert isinstance(out_size, int) and out_size > 0, "out_size must be positive int"
self.edge_embedding_layer = ms.nn.SequentialCell(
ms.nn.Dense(edge_feat_size, hidden_size),
ShiftedSoftplus(),
ms.nn.Dense(hidden_size, hidden_size),
ShiftedSoftplus()
)
self.node_embedding_layer = ms.nn.Dense(node_feat_size, hidden_size)
self.out_embedding_layer = ms.nn.SequentialCell(
ms.nn.Dense(hidden_size, out_size),
ShiftedSoftplus()
)
# pylint: disable=arguments-differ
def construct(self, x, edge_feats, g: Graph):
"""
Construct function for CFConv.
"""
g.set_vertex_attr({"hv": self.node_embedding_layer(x)})
g.set_edge_attr({"he": self.edge_embedding_layer(edge_feats)})
for v in g.dst_vertex:
v.h = g.sum([s.hv * e.he for s, e in v.inedges])
return self.out_embedding_layer([v.h for v in g.dst_vertex])