Skip to main content

Table 3 SMAP results for the bundled embeddings

From: Using bundle embeddings to predict daily cortisol levels in human subjects

 

Night bundle

Day bundle

ID

dim

n

nrmsemin

∆nrmse

θ

dim

n

nrmsemin

∆nrmse

θ

1

1

54

0.799

0.001

0.2

2

52

1.254

0.009a

0.2

2

1

57

0.877

0.000

0.0

3

53

1.246

0.005

0.2

3

1

56

0.454

0.002

0.2

1

56

1.505

0.000

0.0

4

2

55

0.802

0.006a

0.2

2

56

1.455

0.000

0.0

5

2

58

0.329

0.001

0.2

1

59

2.353

0.000

0.0

6

1

57

0.842

0.000

0.0

7

40

1.620

0.004

0.2

7

1

55

0.578

0.024b

0.6

3

51

1.307

0.033

0.4

8

1

56

0.680

0.001

0.2

3

52

0.744

0.005b

0.4

9

1

57

0.722

0.046b

0.6

3

56

1.136

0.069b

0.6

10

2

53

0.625

0.006b

0.4

1

55

2.292

0.000

0.0

  1. Note: ID = subject ID, dim = dimension of the embedding, n = number of vectors in the embedding, nrmsemin = minimum of the NRMSE, Δnrmse difference between the minimal NRMSE (θ > = 0) and the NRMSE at θ = 0 (the global linear case), θ = value of θ where the NRMSE was minimal
  2. aSignificant at the 0.05 level
  3. bSignificant at the 0.01 level
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us