Skip to main content

Table 4 Benchmark results of the RL-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

80.96

80.17

77.44

0.81

41.32 = 2076

2

99.52

91.53

10.60

0.66

66.66 = 644

3

86.73

55.22

51.44

0.80

27.75 = 684

4

98.16

97.65

0.33

0.64

51.04 = 16

Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us