|View source on GitHub|
Makes additional inputs for neighbor features if necessary.
nsl.keras.layers.make_missing_neighbor_inputs( neighbor_config, inputs, weight_dtype=tf.float32 )
neighbor_config: An instance of
configs.GraphNeighborConfigspecifying the number of neighbors and how neighbor features should be named.
inputs: Dictionary of input tensors that may be missing neighbor features. The keys are the features names. See
utils.unpack_neighbor_featuresfor expected names of neighbor features and weights.
tf.Dtypefor neighbors weights. Defaults to
A dictionary of neighbor feature and weight tensors that do not already
inputs. The keys are specified according to