Deep learning models are incredibly successful in analyzing massive quantities of data and accurately forecasting future occurrences. Meteorologists can now fairly accurately forecast broad weather patterns for the next two to three days. However, climate change has increased unexpected extreme weather events such as thunderstorms, hailstorms, and hurricanes. Predicting these unexpected weather phenomena accurately a few hours ahead of time might help people prepare for them, perhaps reducing their effects and negative consequences. Three deep neural networks have recently been constructed by researchers at IRT AESE Saint Exupéry and Météo-France to anticipate oncoming precipitation.