This dataset accompanies the paper, "Can Deep Learning Predict Complete Ruptures in Numerical Megathrust Faults?". The directory contains full python codes used to create Convolutional Neural Networks and Long Short Term Memory Recurrent Neural Networks, along with the raw data used for training, testing, and validation. Since stochastic initialization of weights in the training process will result in slightly different parameterization of models trained with identical hyperparameters, we have also included the trained models we presented in the paper with this dataset as well.