arXiv (Cornell University)· 2019· Preprint
PyTorch: An Imperative Style, High-Performance Deep Learning Library
- 16,191citations
- 2019year
Short summary
PyTorch demonstrates that deep learning frameworks can achieve both usability and high performance, offering an imperative, Pythonic style that simplifies debugging and integration while efficiently utilizing hardware accelerators like GPUs.
AI-generated from the title and abstract; the full text is not read.
Key points
- PyTorch combines an imperative, Pythonic programming style with high performance.
- The framework supports easy debugging and integration with scientific computing libraries.
- It efficiently utilizes hardware accelerators such as GPUs.
- Performance is demonstrated through benchmarks of individual subsystems and the overall library.
AI-generated from the title and abstract; the full text is not read.
Abstract
Deep learning frameworks have often focused on either usability or speed, but not both. PyTorch is a machine learning library that shows that these two goals are in fact compatible: it provides an imperative and Pythonic programming style that supports code as a model, makes debugging easy and is consistent with other popular scientific computing libraries, while remaining efficient and supporting hardware accelerators such as GPUs. In this paper, we detail the principles that drove the implementation of PyTorch and how they are reflected in its architecture. We emphasize that every aspect of PyTorch is a regular Python program under the full control of its user. We also explain how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance. We demonstrate the efficiency of individual subsystems, as well as the overall speed of PyTorch on several common benchmarks.
The authors' abstract, as published at the source. arXiv (Cornell University), 2019 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Hardware and Architecture
Hardware and ArchitectureComputer Science