A Deep Neural Network for Gastric Cancer Prognosis Prediction Based on Biological Information Pathways

Table 1

GCS-NET network model parameters.

Layer (type)	Output shape	Param	Connected to
Inputs (InputLayer)	(none, 34380)	0
h0 (Diagonal)	(none, 11460)	45840	Inputs[0][0]
dropout_0 (Dropout)	(none, 11460)	0	h0[0][0]
h1 (SparseTF)	(none, 1061)	22081	dropout_0[0][0]
dropout_1 (Dropout)	(none, 1061)	0	h1[0][0]
h2 (SparseTF)	(none, 447)	1512	dropout_1[0][0]
dropout_2 (Dropout)	(none, 447)	0	h2[0][0]
h3 (SparseTF)	(none, 147)	594	dropout_2[0][0]
dropout_3 (Dropout)	(none, 147)	0	h3[0][0]
h4 (SparseTF)	(none, 26)	174	dropout_3[0][0]
o_linear1 (Dense)	(none, 1)	11461	h0[0][0]
o_linear2 (Dense)	(none, 1)	1062	h1[0][0]
o_linear3 (Dense)	(none, 1)	448	h2[0][0]
o_linear4 (Dense)	(none, 1)	148	h3[0][0]
o_linear5 (Dense)	(none, 1)	27	h4[0][0]
Total		83347