Data-Driven Quality Prediction of Batch Processes Based on Minimal-Redundancy-Maximal-Relevance Integrated Convolutional Neural Network

Table 3

Training RMSE and R² of different methods.

Methods	SVR	BP	AE	SAE	CNN	CNN-L1	mRMR-CNN
RMSE	0.0571	0.0297	0.0270	0.0397	0.0246	0.0293	0.0232
R^2	0.9560	0.9881	0.9901	0.9788	0.9918	0.9884	0.9927

Bold indicates the lowest RMSE value and highest R² value.