Table 2 Comparing DGCNN models for prediction of SOC

From: DGCNN approach links metagenome-derived taxon and functional information providing insight into global soil organic carbon

Estimator

Large-graph

Mid-graph

Small-graph

Test MAE

41.615

41.518

27.194

Train MAE

41.589

40.574

33.914

Mean MAE CV

41.416

41.841

34.081

SD MAE CV

3.638

5.702

3.540

  1. DGCNN model performance comparison between our “Large-graph”, “Mid-graph” and “Small-graph” examples for a range of metrics including Mean Absolute Error (MAE) on the training dataset (Train MAE), MAE on the held-out test set (Test MAE), the mean and standard deviation of the MAE value after five-fold CV (Mean MAE CV, SD MAE CV).
Back to article page