Deep-Learning-Based Cancer Profiles Classification Using Gene Expression Data Profile

Table 2

Proposed classifier input parameters.

Input parameters allocated	Proposed classifier best
	Value analysis
Train size = 9000	Train size = 9000
Test size = 1000	Test size = 1000
nummular = 1	nummular = 0
dropout = False	dropout = False
train_valid_split = 0.5	train_valid_split = 0.5
channels = 4	channels = 32
patience = 10	patience = 10
Weight decay = 0.00000001	Weight decay = 0.00000001
Batch size = 10	Batch size = 10