Major update version of AI library 'PyTorch' used in image generation AI 'PyTorch 2.0' official release

Version 2.0 of the machine learning library `` PyTorch '' for Python, which is also used by famous AI such as the image generation AI `` Stable Diffusion '', has been officially released. PyTorch 2.0 is significantly faster while maintaining backwards compatibility.

PyTorch 2.0: Our next generation release that is faster, more Pythonic and Dynamic as ever | PyTorch
https://pytorch.org/blog/pytorch-2.0-release/

Release PyTorch 2.0: Our next generation release that is faster, more Pythonic and Dynamic as ever pytorch/pytorch GitHub
https://github.com/pytorch/pytorch/releases/tag/v2.0.0

PyTorch is an open source machine learning library for Python that is used in many AI projects. Meta was responsible for the initial development of PyTorch, but in September 2022, the PyTorch project became independent from Meta and managed under the Linux Foundation.

AI framework 'PyTorch', which is also used for Stable Diffusion of image generation AI, moves to Linux Foundation independently from Meta-GIGAZINE

Development of PyTorch is progressing even after the change in management system, and PyTorch 2.0 was officially released on March 15, 2023. The development team appeals that '100% backward compatibility is maintained', 'immediate performance after introduction', and 'significant performance improvement'.

We're excited to announce the release of PyTorch 2.0!

This version includes:
⚙️ 100% backward compatible
???? Out of the box performance
???? Significant speed improvements

Learn more ???? https://t.co/8vV61atP6E pic.twitter.com/NQQEEUUJNl

—PyTorch (@PyTorch) March 15, 2023

The main improvements in PyTorch 2.0 are:

・Learning time is faster with the new compiler 'torch.compile'
・'TorchInductor' makes it possible to execute optimal processing according to the type of GPU and CPU
・Equipped with the stable version of 'Accelerated Transformers (former name: Better Transformer)'
・Compatible with Apple's ' Metal Performance Shaders ' and supports GPU acceleration on Mac
・Improved inference performance in ' AWS Graviton ' of Amazon Web Services (AWS)

PyTorch 2.0 can be used on Windows, macOS, and Linux, and the installation method for each environment is summarized on the following page.

Start Locally | PyTorch
https://pytorch.org/get-started/