Deep learning methods are supported by a variety of software packages and tools written in a number of different programming languages. The Python language is especially popular for implementing deep learning frameworks, and widely used tools such as PyTorch and TensorFlow provide expressive application programming interfaces (APIs) written in Python that leverage numerically efficient compiled libraries written in other languages. Additional packages and libraries contribute to a software ecosystem that supports the organization, manipulation, and visualization of data and predictions from deep learning pipelines.