Fig. 1
From: GCNCPR-ACPs: a novel graph convolution network method for ACPs prediction
The comparison of the performance of our proposed GCNCPR-ACPs with other state-of-the-art predictors. A The effects of the learning rate on the performance of the tenfold cross-validation results of the proposed GCNCPR-ACPs and the existing prediction models. B The effects of the learning rate on the performance of the independent test results of the proposed GCNCPR-ACPs and the existing prediction models. C The effects of the number of layers on the performance of the tenfold cross-validation results of the proposed GCNCPR-ACPs and the existing prediction models. D The effects of the number of layers on the performance of the independent test results of the proposed GCNCPR-ACPs and the existing prediction models. E The effects of the assign ratio on the performance of the tenfold cross-validation results of the proposed GCNCPR-ACPs and the existing prediction models. F The effects of the assign ratio on the performance of the independent test results of the proposed GCNCPR-ACPs and existing prediction models