Skip to main content

Table 5 Benchmark results of the CRLV-biased-G: best epoch

From: Conditional reduction of the loss value versus reinforcement learning for biassing a de-novo drug design generator

Objective

Validity

Novelty

Uniqueness

intDiv

Desirability

1

66.33

86.56

81.60

0.82

68.11 = 3192

2

32.42

97.00

96.43

0.81

42.96 = 1302

3

62.45

88.69

77.48

0.82

22.46 = 964

4

67.21

84.88

74.40

0.80

2.52 = 107

Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us