The past several decades have seen an exponential increase in the volume of available seismic data, and with it has come the need to develop fast, automatic earthquake detection, and location algorithms. Some of the most recent and promising tools come from the field of machine learning. In this study, we combine a recent seismic detection and location method with neural network classification and analyze 4 months of continuous data recorded by a network of 76 stations in northern California. While these approaches have been used separately, our implementation is unique in that it is not constrained by source templates and avoids user-defined detection thresholds. In particular, we partition our data set into 234,240, 3-min long time windows with 75% overlap. For each time window, we create a 3D image that captures information about the coherence of the seismic wavefield. We then devise four features as input and train a neural network classifier to predict which time windows in the data set are likely to contain regional seismic events. These features include the second and fourth Hu image moments computed from 2D cross sections of our 3D images and statistical p values that quantify the probability of observing network-wide power-spectral density values at 0.2 and 0.5 s. Our neural network model predicts that 2,522 time windows contain seismic events, from which we locate 1,192 unique events. Plain Language Summary Whereas seismic data sets were relatively sparse in the past, modern data sets may be as large as several hundred terabytes. Given that researchers are often interested in identifying and studying seismic events (i.e., earthquakes) or their effects, and that these events are hidden within large and noisy data sets, they require tools to automatically and quickly search through large volumes of data to find what they are interested in. In fact, this kind of problem is perfectly suited to machine learning tasks, and this is one reason why machine learning has been so widely adopted in seismology. In this study, we take a reasonably large data set (4 months of data recorded by 76 seismic stations) and use a machine learning algorithm known as a neural network to predict time periods in the data set that are likely to contain earthquakes. Using this approach, we identify 1,192 earthquakes as well as provide location and origin time estimates for them.