Function set_node_lazy_features

python/dgl/dataloading/base.py:23–55

Assign lazy features to the ``ndata`` of the input graph for prefetching optimization. When used in a :class:`~dgl.dataloading.Sampler`, lazy features mark which data should be fetched before computation in model. See :ref:`guide-minibatch-prefetching` for a detailed explanation. I

(g, feature_names)

def set_node_lazy_features(g, feature_names):
    """Assign lazy features to the ``ndata`` of the input graph for prefetching optimization.

    When used in a :class:`~dgl.dataloading.Sampler`, lazy features mark which data
    should be fetched before computation in model. See :ref:`guide-minibatch-prefetching`
    for a detailed explanation.

    If the graph is homogeneous, this is equivalent to:

    .. code:: python

       g.ndata.update({k: LazyFeature(k, g.ndata[dgl.NID]) for k in feature_names})

    If the graph is heterogeneous, this is equivalent to:

    .. code:: python

        for type_, names in feature_names.items():
            g.nodes[type_].data.update(
                {k: LazyFeature(k, g.nodes[type_].data[dgl.NID]) for k in names})

    Parameters
    ----------
    g : DGLGraph
        The graph.
    feature_names : list[str] or dict[str, list[str]]
        The feature names to prefetch.

    See also
    --------
    dgl.LazyFeature
    """
    return _set_lazy_features(g.nodes, g.ndata, feature_names)


def set_edge_lazy_features(g, feature_names):