Skip to main content
. 2021 Dec 1;6(49):33864–33873. doi: 10.1021/acsomega.1c05145

Table 1. Sample Statistical Results of Transformer-Based and GRU-Based Generators.

  pretrained model
optimized model (aver reward = 15)
  transformer GRU transformer GRU
invalid 1.694% ± 0.102% 4.368% ± 0.361% 1.170% ± 0.052% 2.690% ± 0.167%
redundant 0.075% ± 0.022% 0.038% ± 0.015% 30.273% ± 0.525% 33.828% ± 0.616%
retained 98.231% ± 0.115% 95.594% ± 0.351% 68.557% ± 0.552% 63.482% ± 0.456%