Skip to main content
. 2018 Jun 21;9:859. doi: 10.3389/fpls.2018.00859

Table 3.

Test accuracies and root mean square errors (RMSEs) of trained recurrent neural network (RNN) algorithms.

Type of RNN Test accuracy (R2) Test RMSE
Long short-term memory (LSTM) 0.72 0.08
Gated recurrent unit (GRU) 0.68 0.09
Multi-layered LSTM 0.70 0.08
Multi-layered GRU 0.68 0.09