Method dstdata

python/dgl/heterograph.py:2090–2154

Return a node data view for setting/getting destination node features. Let ``g`` be a DGLGraph. If ``g`` is a graph of a single destination node type, ``g.dstdata[feat]`` returns the destination node feature associated with the name ``feat``. One can also set a destination n

(self)

    @property
    def dstdata(self):
        """Return a node data view for setting/getting destination node features.

        Let ``g`` be a DGLGraph. If ``g`` is a graph of a single destination node type,
        ``g.dstdata[feat]`` returns the destination node feature associated with the name
        ``feat``. One can also set a destination node feature associated with the name
        ``feat`` by setting ``g.dstdata[feat]`` to a tensor.

        If ``g`` is a graph of multiple destination node types, ``g.dstdata[feat]`` returns a
        dict[str, Tensor] mapping destination node types to the node features associated with
        the name ``feat`` for the corresponding type. One can also set a node feature
        associated with the name ``feat`` for some destination node type(s) by setting
        ``g.dstdata[feat]`` to a dictionary as described.

        Notes
        -----
        For setting features, the device of the features must be the same as the device
        of the graph.

        Examples
        --------
        The following example uses PyTorch backend.

        >>> import dgl
        >>> import torch

        Set and get feature 'h' for a graph of a single destination node type.

        >>> g = dgl.heterograph({
        ...     ('user', 'plays', 'game'): (torch.tensor([0, 1]), torch.tensor([1, 2]))})
        >>> g.dstdata['h'] = torch.ones(3, 1)
        >>> g.dstdata['h']
        tensor([[1.],
                [1.],
                [1.]])

        Set and get feature 'h' for a graph of multiple destination node types.

        >>> g = dgl.heterograph({
        ...     ('user', 'plays', 'game'): (torch.tensor([1, 2]), torch.tensor([1, 2])),
        ...     ('user', 'watches', 'movie'): (torch.tensor([2, 2]), torch.tensor([1, 1]))
        ... })
        >>> g.dstdata['h'] = {'game': torch.zeros(3, 1), 'movie': torch.ones(2, 1)}
        >>> g.dstdata['h']
        {'game': tensor([[0.], [0.], [0.]]),
         'movie': tensor([[1.], [1.]])}
        >>> g.dstdata['h'] = {'game': torch.ones(3, 1)}
        >>> g.dstdata['h']
        {'game': tensor([[1.], [1.], [1.]]),
         'movie': tensor([[1.], [1.]])}

        See Also
        --------
        nodes
        ndata
        dstnodes
        """
        if len(self.dsttypes) == 1: