Space matters. In the world of data science and machine learning, the size of the dataset is often the biggest determining factor in choosing a model. And in the world of deployment, it is the size of the model that determines whether or not it can be used on a mobile device. Compression can help to solve both problems by reducing the size of datasets and models. This opens up new avenues for storage, transmission, and training/inference.
The Python Compression library is a multi-platform, efficient and easy to use Python library to compress, decompress and extract compressed files. Developers tend to use some of the following open source libraries: Requests - A simple, yet elegant HTTP library, Deep-Compression-AlexNet - Deep Compression on AlexNet, nn-compression - A Pytorch implementation of Neural Network Compression. The entire list of open source libraries are provided below.