mindspore_gl.nn.Set2Set

class mindspore_gl.nn.Set2Set(input_size, num_iters, num_layers)

Sequence to sequence for sets. From the paper Order Matters: Sequence to sequence for sets .

For each subgraph in the batched graph, compute:

\[ \begin{align}\begin{aligned}\begin{split}q_t = \mathrm{LSTM} (q^*_{t-1}) \\\end{split}\\\begin{split}\alpha_{i,t} = \mathrm{softmax}(x_i \cdot q_t) \\\end{split}\\\begin{split}r_t = \sum_{i=1}^N \alpha_{i,t} x_i\\\end{split}\\q^*_t = q_t \Vert r_t\end{aligned}\end{align} \]

Parameters

• input_size (int) – dim for input node features.

• num_iters (int) – number of iterations.

• num_layers (int) – number of layers.

Inputs:

• x (Tensor) - The input node features to be updated. The shape is \((N, D)\) where \(N\) is the number of nodes, and \(D\) is the feature size of nodes.

• g (BatchedGraph) - The input graph.

Outputs:

• x (Tensor) - The output representation for graphs. The shape is \((2, D_{out})\) where \(D_{out}\) is the double feature size of nodes

Raises

TypeError – If input_size or num_iters or num_layers is not an int.

Supported Platforms:

Ascend GPU

Examples

>>> import numpy as np
>>> import mindspore as ms
>>> from mindspore_gl.nn import Set2Set
>>> 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 = Set2Set(4, 3, 2)
>>> ret = net(node_feat, *batched_graph_field.get_batched_graph())
>>> print(ret.shape)
(2, 8)