Fig. 9From: Real-time earthquake magnitude estimation via a deep learning network based on waveform and text mixed modalSpatial distribution of the magnitude prediction errors of the proposed model in the subset region when the \({t}_{1}\) is at 1 and 3 s. The top and bottom subfigures indicate the results when \({t}_{1}\) is at 1 and 3 s, respectively. The left to right subfigures indicate the prediction errors of the earthquakes from different depth segments. The gray triangles represent seismometers. The squares and circles represent the earthquakes located in the inland and offshore, respectively. Different colors indicate the error levelsBack to article page