<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="3.7.3">Jekyll</generator><link href="https://pytorch.org/feed.xml" rel="self" type="application/atom+xml" /><link href="https://pytorch.org/" rel="alternate" type="text/html" /><updated>2020-12-02T01:29:38-08:00</updated><id>https://pytorch.org/</id><title type="html">PyTorch Website</title><subtitle>Scientific Computing...</subtitle><author><name>Facebook</name></author><entry><title type="html">Prototype Features Now Available - APIs for Hardware Accelerated Mobile and ARM64 Builds</title><link href="https://pytorch.org/blog/prototype-features-now-available-apis-for-hardware-accelerated-mobile-and-arm64-builds/" rel="alternate" type="text/html" title="Prototype Features Now Available - APIs for Hardware Accelerated Mobile and ARM64 Builds" /><published>2020-11-12T00:00:00-08:00</published><updated>2020-11-12T00:00:00-08:00</updated><id>https://pytorch.org/blog/prototype-features-now-available-apis-for-hardware-accelerated-mobile-and-arm64-builds</id><content type="html" xml:base="https://pytorch.org/blog/prototype-features-now-available-apis-for-hardware-accelerated-mobile-and-arm64-builds/">&lt;p&gt;Today, we are announcing four PyTorch prototype features. The first three of these will enable Mobile machine-learning developers to execute models on the full set of hardware (HW) engines making up a system-on-chip (SOC). This gives developers options to optimize their model execution for unique performance, power, and system-level concurrency.&lt;/p&gt;

<li>If you want to build on Windows, Visual Studio 2017 14.11 toolset and NVTX are also needed. Especially, for CUDA 8 build on Windows, there will be an additional requirement for VS 2015 Update 3 and a patch for it. The details of the patch can be found out <a href="https://support.microsoft.com/en-gb/help/4020481/fix-link-exe-crashes-with-a-fatal-lnk1000-error-when-you-use-wholearch">here</a>.</li>

3. If you want to build on Windows, Visual Studio 2017 14.11 toolset and NVTX are also needed. Especially, for CUDA 8 build on Windows, there will be an additional requirement for VS 2015 Update 3 and a patch for it. The details of the patch can be found out [here](https://support.microsoft.com/en-gb/help/4020481/fix-link-exe-crashes-with-a-fatal-lnk1000-error-when-you-use-wholearch).