Over the last year, Microsoft has been developing GPU-accelerated training for machine learning models on its Windows Subsystem for Linux (WSL) service. After the last year of previewing, Microsoft is now launching the project as TensorFlow-DirectML.
As the name suggests, TensorFlow-DirectML is a fork including Google’s TensorFlow machine learning platform and Microsoft’s DirectML API.
DirectML offers GPU acceleration tools for machine learning tasks. DirectML works across hardware and drivers on DirectX 12 GPUs on AMD, Qualcomm, Nvidia, and Intel chips. With the TensorFlow-DirectML, Microsoft is providing a hardware acceleration platform that can be accessed by multiple vendors to train machine learning models.
Microsoft says the open source project is now fully available via GitHub.
“Today we’re excited to exit preview and announce our first generally consumable package of TensorFlow-DirectML! We encourage you to use TensorFlow-DirectML whether you’re a student learning or a professional developing machine learning models for production. Read on for more details on how we invested and where we’re headed next.”
To help release the best version of TensorFlow-DirectML, Microsoft sought user feedback and worked with partners such as AMD, Intel, and Nvidia. Microsoft says the following requirements are needed to run the service.
Requirements
Windows 10
- Windows 10 Version 1709, 64-bit (Build 16299 or higher)
- Python x86-64 3.5, 3.6, or 3.7
- One of the following supported GPUs:
- AMD Radeon R5/R7/R9 2xx series or newer
- Intel HD Graphics 5xx or newer
- NVIDIA GeForce GTX 9xx series GPU or newer
Note: Python 3.8 or newer is not currently supported. To use the official PyPi packages, the CPython interpreter is required. NumPy 1.19.4 is requires the KB4598291 to properly work on Windows.
Windows Subsystem for Linux
- Windows 10 Insider Preview, 64-bit (Build 20150 or higher)
- Python x86-64 3.5, 3.6, or 3.7
- One of the following supported GPUs:
- AMD Radeon R5/R7/R9 2xx series or newer, and 20.20.01.05 WSL driver
- Intel HD Graphics 6xx or newer, and 28.20.100.8322 WSL driver
- NVIDIA GeForce GTX 9xx series GPU or newer, and 460.20 WSL driver
Note: Python 3.8 or newer is not currently supported. To use the official PyPi packages, the CPython interpreter is required.
The new TensorFlow-DirectML is available on GitHub here.
Last Updated on February 14, 2022 8:20 pm CET by Luke Jones
