Research Article

A Multichannel 2D Convolutional Neural Network Model for Task-Evoked fMRI Data Classification

Figure 8

Mean training loss (in solid lines) and mean validation loss (in dashed lines) for 3D CNN (red) and M2D CNN (blue). The standard deviation is indicated by the shadow area in the image. (a) 2000 samples. (b) 5000 samples.
(a)
(b)