TensorFlow 2.15 发布，带来更简单的 NVIDIA CUDA 库 Linux 安装方法，Windows x64 和 x86 的 oneDNN CPU 性能优化，tf.function 类型全面可用，Clang 17.0.1 升级等更新。新版本简化了 CUDA 库的安装流程，提升了 Windows 平台的 CPU 性能，并增强了 tf.function 的功能。此外，TensorFlow PIP 包已使用 Clang 17 和 CUDA 12.2 构建，以提升 NVIDIA Hopper 架构 GPU 的性能。

🔧 TensorFlow 2.15 发布，简化了 Linux 平台上 NVIDIA CUDA 库的安装方法。用户只需安装 NVIDIA 驱动后，即可通过 pip install tensorflow[and-cuda] 命令安装所需的 CUDA 库依赖，无需预先安装其他 CUDA 包。CUDA 版本已升级至 12.2。

🚀 Windows x64 和 x86 版本的 TensorFlow 2.15 启用了 oneDNN 优化，默认在 X86 CPU 上运行。用户可通过设置环境变量 TF_ENABLE_ONEDNN_OPTS 来启用或禁用该优化，以适应不同需求。

🔗 tf.function 类型全面可用，引入了 tf.types.experimental.TraceType 和 tf.types.experimental.FunctionType，支持自定义输入的 Tensor 分解和类型转换。同时，tf.types.experimental.AtomicFunction 提供了更快的 Python TF 计算方式，但不支持梯度计算。

🛠️ TensorFlow PIP 包已升级至 Clang 17.0.1 和 CUDA 12.2，以提升 NVIDIA Hopper 架构 GPU 的性能。未来 Clang 17 将成为 TensorFlow 的默认 C++ 编译器，建议用户在从源码构建 TensorFlow 时升级编译器。

Posted by the TensorFlow team

TensorFlow 2.15 has been released! Highlights of this release (and 2.14) include a much simpler installation method for NVIDIA CUDA libraries for Linux, oneDNN CPU performance optimizations for Windows x64 and x86, full availability of tf.function types, an upgrade to Clang 17.0.1, and much more! For the full release note, please check here.

TensorFlow Core

NVIDIA CUDA libraries for Linux

The tensorflow pip package has a new, optional installation method for Linux that installs necessary NVIDIA CUDA libraries through pip. As long as the NVIDIA driver is already installed on the system, you may now run pip install tensorflow[and-cuda] to install TensorFlow's NVIDIA CUDA library dependencies in the Python environment. Aside from the NVIDIA driver, no other pre-existing NVIDIA CUDA packages are necessary. In TensorFlow 2.15, CUDA has been upgraded to version 12.2.

oneDNN CPU performance optimizations

For Windows x64 & x86 packages, oneDNN optimizations are now enabled by default on X86 CPUs. These optimizations can be enabled or disabled by setting the environment variable TF_ENABLE_ONEDNN_OPTS to 1 (enable) or 0 (disable) before running TensorFlow. To fall back to default settings, simply unset the environment variable.

tf.function

tf.function types are now fully available.

tf.types.experimental.TraceType now allows custom tf.function inputs to declare Tensor decomposition and type casting support. 

Introducing tf.types.experimental.FunctionType as the comprehensive representation of the signature of tf.function callables. It can be accessed through the function_type property of tf.function’s and ConcreteFunctions. See the tf.types.experimental.FunctionType documentation for more details. 

Introducing tf.types.experimental.AtomicFunction as the fastest way to perform TF computations in Python. This capability can be accessed through the inference_fn property of ConcreteFunctions. (Does not support gradients.) See the tf.types.experimental.AtomicFunction documentation for how to call and use it.

Upgrade to Clang 17.0.1 and CUDA 12.2

TensorFlow PIP packages are now being built with Clang 17 and CUDA 12.2 to improve performance for NVIDIA Hopper-based GPUs. Moving forward, Clang 17 will be the default C++ compiler for TensorFlow. We recommend upgrading your compiler to Clang 17 when building TensorFlow from source.