Speeding up and reducing memory usage for scientific machine learning via mixed precision

Open in new window