PlaidML

PlaidML is a portable tensor compiler. Tensor compilers bridge the gap between the universal mathematical descriptions of deep learning operations, such as convolution, and the platform and chip specific code needed to perform those operations with good performance. Internally, PlaidML makes use of the Tile eDSL [3] to generate OpenCL, OpenGL, LLVM, or CUDA code. It enables deep learning on devices where the available computing hardware is either not well supported or the available software stack contains only proprietary components. For example, it does not require the usage of CUDA or cuDNN on Nvidia hardware, while achieving comparable performance[4].

PlaidML
Original author(s)Vertex.AI
Developer(s)Intel
Initial release20 October 2017 (2017-10-20) [1]
Operating systemLinux,[2]
Mac OS,[2]
Microsoft Windows [2]
Type
  • Tensor compiler
LicenseApache License 2.0 [2]
Websitegithub.com/plaidml/plaidml

PlaidML supports the machine learning libraries Keras, ONNX, and nGraph.

History

In August 2018 Intel acquired Vertex.AI, a startup whose mission statement was “deep learning for every platform”.[5] Intel released PlaidML as free software under to the terms of the Apache Licence (version 2.0) to improve compatibility with nGraph, TensorFlow, and other ecosystem software.

gollark: Currently I'm watching a random YouTube video, looking at Discord, and managing my ebook library.
gollark: Listening to train noises all day is not, personally, my idea of interesting.
gollark: We already have had those.
gollark: I read some rough prediction that humanity has a 1 in 6 risk of being entirely wiped out in the next 100 years.
gollark: Neither of which would be very good, of course.

References
  1. @plaidml (20 October 2017). "Hello world! We're live on GitHub and PyPI. Open source deep learning for any GPU. #OpenCL #Keras github.com/plaidml/plaidml" (Tweet) – via Twitter.
  2. PlaidML Github page
  3. C++ Tile eDSL
  4. https://github.com/plaidml/plaidml/blob/master/README.md
  5. Press statement concerning the acquisition of Vertex.AI
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.