Figure 1 | Scientific Reports

Figure 1

From: Locating induced earthquakes with a network of seismic stations in Oklahoma via a deep learning method

Figure 1

The neural network architecture and network training. (a) The sizes of the input and output data are labeled on the left side of the network architecture, and the depth (channel) of the data is labeled to the right. The images of eight selected channels from some of the layers are displayed. (b) The red box marks the region of interest, and 1,013 cataloged events (ML ≥ 3.0) (blue dots) are selected for the training set. (c) The loss curve during the training is displayed.

Back to article page