Transformer support for PyTorch with DirectML is here!

December 6, 2022 Dec 6, 2022 12/6/22
Adele Parsons
The latest release of PyTorch with DirectML is available today! This release brings support for training popular Transformer Models like GPT2, BERT, and Detection Transformers. To get started with training Transformer Models using PyTorch with DirectML, you can find a new sample on the DirectML GitHub. The sample covers training a PyTorch ...

Real-Time Image Blurring & DirectX Resource Binding in the Windows ML Samples Gallery

September 29, 2022 Sep 29, 2022 09/29/22
authorsLinnea May, Numfor Mbiziwo-Tiapo
Last November, we announced the release of the Windows ML Samples Gallery. The newest release of the gallery adds 4 new samples, 2 of which are discussed in detail below.    Real-Time Background Image Blurring  Linnea May   Real-time machine learning inference is a hot topic in ML, with applications such as real-time object ...
Comments are closed.0 2

DirectML Plugin for TensorFlow 2 is here

June 22, 2022 Jun 22, 2022 06/22/22
Jonah Dykhuizen
Last October we announced our generally consumable DirectML backend for TensorFlow 1.15. Since then, our team worked to bring machine learning training acceleration to TensorFlow 2. Today, we are happy to announce the release of our DirectML PluggableDevice preview package for TensorFlow on PyPI. Alongside our release, we are excited to ...
Comments are closed.0 1

PyTorch-DirectML: Preview Release 3

May 25, 2022 May 25, 2022 05/25/22
Adele Parsons
The Windows AI Platform is excited to announce the Third Preview release of PyTorch-DirectML; a hardware accelerated backend for training PyTorch models on any DirectX12 GPU on Windows and the Windows Subsystem for Linux (WSL). This preview release includes expanded support for computer vision models from the Torchvision library. On the ...
Comments are closed.0 0

PyTorch-DirectML: Preview Release 2

February 25, 2022 Feb 25, 2022 02/25/22
Adele Parsons
A few months ago, we released the first preview of PyTorch-DirectML: a hardware accelerated backend for training PyTorch models on any DirectX12 GPU on Windows and the Windows Subsystem for Linux (WSL). Today, we are releasing the Second Preview with significant performance improvements and greater coverage for computer vision models. ...
Comments are closed.0 0

Introducing PyTorch-DirectML: Train your machine learning models on any GPU

October 21, 2021 Oct 21, 2021 10/21/21
Adele Parsons
The Windows AI team is excited to announce the first preview of DirectML as a backend to PyTorch for training ML models! This release is our first step towards unlocking accelerated machine learning training for PyTorch on any DirectX12 GPU on Windows and the Windows Subsystem for Linux (WSL). In order for you to take advantage of DirectML ...

Train your machine learning models on any GPU with TensorFlow-DirectML

September 9, 2021 Sep 9, 2021 09/9/21
Clarke Rahrig
Learn about the first generally consumable package of TensorFlow-DirectML. TensorFlow-DirectML improves the experience and performance of model training through GPU acceleration on the breadth of Windows devices by working across different hardware vendors and supporting the needs of students’ and professionals’ workflows.