Fig. 10.
Permutation importance for the neural network regressor, normalised by L⋆. The average importance over the four train-test split models are shown, and the standard deviation is given as error bars. The importance of a feature is the RMSE of the model after randomizing that feature, minus the RMSE of the unpermuted predictions.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.