Deep convolutional neural network with an Encoder-Decoder architecture applied to learn TMS E-field. The numbers underneath convolutional (C1_1, C1_2, …, C9_2) and deconvolutional (D1, D2, D3, and D4) layers indicate their corresponding numbers of kernels, with a stride of 1 or 2 for downsampling or upsampling. The kernel size in all layers is set to 3×3×3.