# 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.
# ============================================================================
"""Sort Pooling Layer."""
import mindspore as ms
from mindspore_gl import BatchedGraph
from .. import GNNCell
[文档]class SortPooling(GNNCell):
r"""
Apply sort pooling to the nodes in the graph.
From the paper `End-to-End Deep Learning Architecture for Graph Classification <https://muhanzhang.github.io/papers/AAAI_2018_DGCNN.pdf>`_ .
The sorting pool first sorts the node features in ascending order along the feature dimension,
and then selects the ranking features of top-k nodes (sorted by the maximum value of each node).
Args:
k (int): Number of nodes to keep per graph.
Inputs:
- **x** (Tensor) - The input node features to be updated. The shape is :math:`(N, D)`
where :math:`N` is the number of nodes, and :math:`D` is the feature size of nodes.
- **g** (BatchedGraph) - The input graph.
Outputs:
- **x** (Tensor) - The output representation for graphs. The shape is :math:`(2, D_{out})`
where :math:`D_{out}` is the double feature size of nodes.
Raises:
TypeError: If `k` is not an int.
Supported Platforms:
``Ascend`` ``GPU``
Examples:
>>> import numpy as np
>>> import mindspore as ms
>>> from mindspore_gl.nn import SortPooling
>>> from mindspore_gl import BatchedGraphField
>>> n_nodes = 7
>>> n_edges = 8
>>> src_idx = ms.Tensor([0, 2, 2, 3, 4, 5, 5, 6], ms.int32)
>>> dst_idx = ms.Tensor([1, 0, 1, 5, 3, 4, 6, 4], ms.int32)
>>> ver_subgraph_idx = ms.Tensor([0, 0, 0, 1, 1, 1, 1], ms.int32)
>>> edge_subgraph_idx = ms.Tensor([0, 0, 0, 1, 1, 1, 1, 1], ms.int32)
>>> graph_mask = ms.Tensor([1, 1], ms.int32)
>>> batched_graph_field = BatchedGraphField(src_idx, dst_idx, n_nodes, n_edges, ver_subgraph_idx,
... edge_subgraph_idx, graph_mask)
>>> node_feat = np.random.random((n_nodes, 4))
>>> node_feat = ms.Tensor(node_feat, ms.float32)
>>> net = SortPooling(k=2)
>>> ret = net(node_feat, *batched_graph_field.get_batched_graph())
>>> print(ret.shape)
(2, 8)
"""
def __init__(self, k):
super().__init__()
if k <= 0 or not isinstance(k, int):
raise ValueError("k must be positive int")
self.k = k
# pylint: disable=arguments-differ
def construct(self, x, g: BatchedGraph):
"""
Construct function for SortPooling.
"""
x, _ = ms.ops.Sort()(x)
ret, _ = g.topk_nodes(x, self.k, -1)
ret = ms.ops.Reshape()(ret, (-1, self.k * ms.ops.Shape()(x)[-1]))
return ret