Member-only story
PyTorch 1.13 Officially Released: CUDA upgrade, integration of multiple libraries, M1 chip support
Share PyTorch-related technical updates, tutorial applications
Content guide
Recently, the PyTorch team announced the release of Pytorch 1.13 on the official blog. This article will introduce the four highlights of the new version in detail.
According to the official introduction, PyTorch 1.13 includes the stable version of BetterTransformer, and no longer supports CUDA 10.2 and 11.3, and completed the migration to CUDA 11.6 and 11.7. In addition, the Beta version also adds support for the Apple M1 chip and functorch.
functorch is a library in PyTorch
Provides composable vmap (vectorized) and autodiff conversions
Summary of unmissable highlights of PyTorch 1.13:
1. The BetterTransformer feature set (feature set) supports the general Transformer model during the inference process, and can perform fastpath execution without modifying the model.
