Source: Deep Learning on Medium
[Paper Highlights]Why PyTorch was built (not in a day, obviously)?
The title is an adaptation of an old adage — ‘Rome was not built in a day’ which stands for the belief that it takes time to build great things.
PyTorch is a Machine Learning library built for usability and speed. It provides an imperative and Pythonic programming style that supports code as a model and makes debugging easy. It is consistent with other popular scientific computing libraries and supports hardware accelerators such as GPUs while remaining efficient.
Internally, PyTorch performs immediate execution of dynamic tensor computations with automatic differentiation and GPU acceleration and does so while maintaining performance comparable to the fastest current libraries for deep learning.