Deep learning is a rapidly evolving field as new and exciting research is released. The community is beginning to shift from a model centric approach, which focuses on the neural network algorithm, to a more data centric approach, which focuses on the training data and the quality of the labels. If you create better labels, the network will learn more efficiently and converge to a better solution in a short amount of time. This is critical when applied to seismic data to better characterize the subsurface.