Fig. 5From: Identifying piRNA targets on mRNAs in C. elegans using a deep multi-head attention networkThe performance comparison of the devised model and the pure CNN model on the wild-type CLASH ground truth split runs. a Learning curves for the pure CNN model. b The comparison between the devised network and the pure CNN model on the test sets in the 30 random split runs. The blue line and red line represent the average results of the devised network and the pure CNN model, respectively. And the colored dashed line describes the variation for the models in the 30 split runsBack to article page