Skip to main content

Table 2 The generated dataset for the resampling training set

From: Multi-cancer early detection based on serum surface-enhanced Raman spectroscopy with deep learning: a large-scale case–control study

Category

HC

BC

LC

TC

CC

GC

EC

Raw data

1896

569

513

220

215

100

38

Test set

569

171

154

66

65

30

11

Training set

1327

398

359

154

150

70

27

Over sampling γ = 0.6

1327

938

765

663

593

541

501

Over sampling γ = 0.85

1327

1165

1080

1023

981

948

921

Over sampling γ = 1

1327

1327

1327

1327

1327

1327

1327