Figure 4
From: Estimating the effective fields of spin configurations using a deep learning technique
Estimating the x, y, and z components of the effective field from the test and experimental data. (a) The estimated effective field when feeding simulated test data, which contains only the z-axis component, to the trained network. (b) The estimated effective field result of feeding the STXM data to the trained network. (c) The estimated effective field result of feeding the MOKE microscopy data to the trained network. The figures on the left have only one axial spin component as input data, and the figures on the right are the estimated result by the trained network.
