Attention-Based Graph Convolutional Network for Zero-Shot Learning with Pre-Training

Algorithm 2

ADGPZ algorithm.

	Input: Graph , Number of nodes N, Input node characteristics X, Pretrained ResNet50 model classifier parameters
	Output: Classifier parameter , Predicted categories of Unseen classes .
(1)	Initializes: the graph convolutional network parameters.
(2)	Change the Graph to a dense Graph , get the adjacency matrix A.
(3)	while not converged do
(4)	Update by equation (4);
(5)	for Attention-layer do
(6)	Update by equation (10);
(7)	Update by equation (9);
(8)	end for
(9)	Loss = LossFunction (, ), Loss Function update by equation (2) or (3);
(10)	Loss.backward;
(11)	end while
(12)	return
(13)	is obtained by using as classifier parameter of classification .